ArticlePDF Available

The behavior of crude oil spot and futures prices around OPEC and SPR announcements: An event study perspective

Authors:

Abstract and Figures

This paper examines the informational efficiency of crude oil spot and futures markets with respect to OPEC conference and U.S. Strategic Petroleum Reserve (SPR) announcements. We employ the event study methodology to examine the abnormal returns in crude oil spot and futures markets around OPEC conference and SPR announcement dates between 1983 and 2008. Our findings regarding OPEC announcements indicate an asymmetry in that only OPEC production cut announcements yield a statistically significant impact with the impact diminishing for longer maturities. We also find that the persistence of returns following OPEC production cut announcements creates substantial excess returns to investors who take long positions on the day following the end of OPEC conferences. In the case of SPR announcements, we find that the government's use of this program initiates a short-run market reaction following the announcement date. Furthermore, our tests of cumulative abnormal returns suggest that the market reacts efficiently to SPR announcements providing support for the use of the strategic reserves as a tool to stabilize the oil market. Our findings have significant policy implications for investors and are useful in designing effective energy policy strategies.
Content may be subject to copyright.
The Behavior of Crude Oil Spot and Futures Prices around OPEC and SPR Announcements: An
Event Study Perspective
Rıza Demirer†* and Ali M. Kutan**
* Southern Illinois University Edwardsville
** Southern Illinois University Edwardsville; The Emerging Markets Group, Sir Cass Business School,
London; The William Davidson Institute, University of Michigan Business School.
Corresponding author: Rıza Demirer. Department of Economics and Finance, Southern Illinois University Edwardsville,
School of Business, Edwardsville, IL 62026-1102. E-mail: rdemire@siue.edu; Tel: 618-650-2939; Fax: 1-618-650-3047.
The Behavior of Crude Oil Spot and Futures Prices around OPEC and SPR Announcements: An
Event Study Perspective
Abstract
This paper examines the informational efficiency of crude oil spot and futures markets with respect to
OPEC conference and U.S. Strategic Petroleum Reserve (SPR) announcements. We employ the event
study methodology to examine the abnormal returns in crude oil spot and futures markets around
OPEC conference and SPR announcement dates between 1983 and 2008. Our findings regarding
OPEC announcements indicate an asymmetry in that only OPEC production cut announcements yield
a statistically significant impact with the impact diminishing for longer maturities. We also find that
the persistence of returns following OPEC production cut announcements creates substantial excess
returns to investors who take long positions on the day following the end of OPEC conferences. In the
case of SPR announcements, we find that the government’s use of this program initiates a short-run
market reaction lasting for about a week following the announcement date. However, our tests of
abnormal returns suggest no statistically significant impact of these announcements on the oil market,
putting doubt on the efficiency of the use of the reserves as a tool to stabilize the oil market.
Keywords: Crude Oil, OPEC, Strategic Petroleum Reserves, Futures Markets, Event Study
2
1. Introduction
This paper uses the event study methodology to study the impact of two oil-related events on crude oil
market activity. The first type of event is OPEC announcements regarding production quotas. Here, we
look at three types of OPEC announcements: increase, no change, and decrease in quotas. The second
type of event is U.S. Strategic Petroleum Reserve (SPR) announcements on crude oil purchases
(releases) from (to) the market. In doing so, we contribute to the literature in several significant ways.
First, we provide evidence from both spot and futures oil markets. Previous studies (summarized
below) do not tend to cover both markets simultaneously. For the latter, we use up to 12-month
maturities to examine whether these events affect more short- or long-end of the market. This is
important for hedging and speculative strategies as informational efficiency of futures markets is
shown to improve hedging performance. Second, unlike previous work, we perform sensitivity
analysis by employing three different measures of abnormal (excess) returns estimated using the
market model, the ARCH model, and the three-factor Fama-French model. Finally, we use a rich
sample period running from1983 through 2008. This period covers important episodes in the crude oil
market history and provides more robust results based on a larger number of events considered.
There are limited published studies examining the impact of OPEC decisions on oil market activity.
Draper (1984) investigates the impact of scheduled and unscheduled (“emergency”) OPEC
announcements on different maturities in the heating oil market using weekly returns. The results
indicate significant differences in pre- and post average weekly returns for regularly scheduled
meetings, but not for unscheduled ones. Our paper is different from this study as we employ different
normal performance models to measure abnormal returns and examine the behavior of cumulative
abnormal return paths in order to test for persistence in excess returns. In addition,we focus on daily
returns, while Draper (1984) calculates weekly returns.
3
Deaves and Krinsky (1992) examine the reaction of oil futures (crude and heating oil) to OPEC
meetings during the period 1970-1990. They classify OPEC meeting outcomes into two categories
(bad and good news) depending on the sign of the abnormal returns on the day following the
announcement. Using the ARCH methodology to model normal performance, they find statistically
and economically significant excess returns following conferences that were associated with good
news. Our paper is different from this study, as we utilize three different models to estimate normal
performance and look at both spot and future returns for different maturities. In a more recent study on
the effect of OPEC meetings on oil prices, Wirl and Kujundzic (2004) find a weak impact that is at
best restricted to meetings that recommend price hikes.
Guidi, Russel and Tarbert (2006) use the event study methodology to examine the effects of OPEC
production increases and decreases on stock returns in the US and UK, as well as on oil prices. They
look at periods of conflict (i.e., Iraq war) and peace during 1986-2004 and find that OPEC production
“cut” decisions have a much bigher impact on spot oil prices than decisions to “increase”, indicating
asymmetry. Our study is different from theirs in significant aspects. First, they focus on oil spot prices,
while we also study future oil prices. Second, they employ the market model in the event study, while
we use different models to measure normal performance. In addition to “increases” or “decreases” in
quotas, we also look at “no change in production quotas” decisions as market participants may also
react to “no news”.
Another important aspect of our study is the inclusion of the events regarding U.S. SPR
announcements. The SPR Program was established in 1977 following the oil embargo of 1973 in
order to prevent the negative effects of a major petroleum supply interruption on economic activity by
having sufficient petroleum reserves. The proponents of the Program argues that the use of the reserves
in case of significant negative oil shocks may act as an effective tool to stabilize the oil market in the
U.S. and other parts of the world. However, others have argued that such stockpiling of oil by
4
governments to deal with short term restrictions on supply imposed by both OPEC or non-OPEC
nations may not be effective (e.g., Taylor and Van Doren, 2005) or that existing stocks are insufficient
to significantly influence price of oil (Considine, 2006). By examining the impact of SPR
announcements on both spot and future oil prices, we try to shed some light on this important current
debate: Is the SPR Program an effective tool to stabilize the market?
2. Data and Event Descriptions
We examine daily spot and futures prices for light sweet crude oil over the period March 1983 through
June 2008. We concentrate on crude oil contracts traded in the U.S. as the SPR program was
established with the goal of stabilizing the oil market in the U.S. Spot prices are measured by cash
prices with the anticipation that cash prices will better reflect actual transactions. In order to examine
the information content of OPEC and SPR announcements for different time horizons, we also
compiled daily NYMEX light sweet crude oil futures settlement prices for a number of maturities into
the future. More specifically, we examine the nearest contract as well as the third and twelfth closest
contracts covering a range of maturities spanning up to one year into the future. Futures return series
are constructed with contract rollover occurring about one week before maturity in most cases.
Table 1 presents several summary statistics for daily spot as well as futures returns for the different
maturities mentioned earlier. As expected, we observe the highest volatility in the spot market with
lower return volatility values as we move towards longer maturities. Observing the highest volatility in
the cash market is not surprising as these prices reflect transactions for immediate delivery of the
underlying asset and tend to be more sensitive to supply and demand changes in the market. This can
also be seen in the extreme values observed in these return series with daily spot returns ranging
between a lowest value of -40.2% and a highest value of 19.86%. High correlation values observed
between spot and futures return series indicate the common risk factors underlying the oil market.
5
In this study, we explore the information content of OPEC announcements as well as SPR
announcements. For this purpose, we examined meeting summaries from the Official Resolutions and
Press Releases published by the OPEC Secretariat and compiled a list of official announcements on
production decisions. OPEC meets twice a year on prescheduled dates for ‘ordinary’ conferences but
they also call for ‘extraordinary’ conferences with short notice. The ministerial meetings are held
occasionally to resolve operational and monitoring problems in the organization; and sometimes they
decide to change production levels. In our analysis, each official press release is considered an event.
Having compiled a list of events, we then classified each OPEC announcement in terms of a
production cut, hike and no change in production levels. As will be explained in the methodology
section, several announcements had to be omitted from the analysis due to the specifics of the event
study methodology. Overall, as reported in Table 2, a total of 63 OPEC meetings have been examined,
of which 17 resulted in a production hike, 21 in a production cut, and 25 in no change in production
levels. Similarly, U.S. government’s announcements on the use of Strategic Petroleum Reserves are
compiled from the Energy Information Administration’s website.1 During the period studied, 15 SPR
related announcements have been made, of which 11 were announcements on the release of crude oil
from the reserves to the market (SPR decrease) and 4 were on the purchase of crude oil from the
market at market prices (SPR increase). Crude oil releases from the SPR have been either in the form
of a sale or a loan to the market (mostly refineries) where purchases have been either in the form of a
direct purchase or payment by oil companies in return for federal leases. Whatever the form of oil
transfer may be, an SPR decrease announcement indicates additional oil supply to the market and an
SPR increase announcement indicates and additional purchase from the market. It is in fact a source of
heated discussions among policy makers whether it is in the best interest of consumers and taxpayers
when the government purchases large quantities of oil from the market only to create upward pressure
in prices. Furthermore, such action makes little sense when the government purchases oil at the already
1 www.eia.doe.gov
6
high market prices when it should actually be selling in order to benefit taxpayers. Therefore, it is an
interesting question both from a research as well as policy making standpoint whether SPR actions
have created the desired market reaction without hurting consumers by artificially inflating market
prices.
3. Methodology
Event study methodology is a widely used analytical tool that has been applied to a wide range of
corporate events in the empirical finance literature. The goal is to assess the information content of an
event such as a merger or an earnings announcement by examining the behavior of abnormal or excess
returns earned on the underlying security around a recurrent relevant event. A large body of literature
has been devoted to the application of this methodology to a number of firm specific and economy
wide events including stock splits, mergers and acquisitions, earnings announcements, international
listings, issues of new securities, and announcements of macro-economics variables. A discussion of
this methodology can be found in Brown and Warner (1980, 1985), Peterson (1989), Thompson
(1995), MacKinlay (1997), and Binder (1998). Although the methodology has been widely applied to
corporate events, its application to non-corporate events has been limited. As mentioned earlier,
Draper(1984) examined the behavior of heating oil futures prices around known events (scheduled
OPEC meetings) and unknown events (special meetings) and found that the market showed a slow
adjustment to information or a possible misestimate of the informational impact. In another study,
Deaves and Krinsky (1992) used the event study methodology to examine the behavior of crude and
heating oil futures prices around OPEC conferences and tested whether the market reacted to these
announcements as prescribed by the efficient market hypothesis. Their analysis of oil prices during the
1980s suggest that if investors took long positions in oil futures on the day following the end of the
conferences associated with good news, they were able to generate substantial positive returns. These
7
excess returns were found to be too high to be explained by transaction costs or risk adjustments. We
extend the analysis in these studies by investigating the events during 1990s and 2000s as well.
3.1. Oil Spot and Futures Excess (Abnormal) Returns
In this study, we use the event study methodology to assess the information content of both OPEC and
SPR announcements using a longer sample period (1983-2008), including more recent data and a
wider range of futures contracts covering a number of maturities. We also perform separate tests using
three different models to measure normal performance. At the heart of an event study is the estimation
of excess or abnormal returns defined as the difference between the actual ex post return of the
underlying asset minus the normal return defined as
)(ctctct RERRA
=
(1)
where Rct is the log return on crude oil on day t and E(Rct) is the normal return. In our study, each
OPEC conference or SPR press release is considered as one event. So, the subscript c indicates the
particular OPEC conference announcement or the SPR announcement that applies to the event period
over which abnormal returns are calculated. A total of 78 events have been compiled (see Table 2). As
mentioned earlier, we perform separate tests in order to examine the behavior of spot returns and three
futures return series, F1, F3, and F12 representing the nearby, third closest and twelfth closest
contracts, respectively. The normal return is defined as the expected return in non event periods
without conditioning on the event taking place. Different models have been proposed in the literature
to estimate normal performance. In this study, we use three separate models to measure normal
performance during non event periods:
1. The Market Model: The market model takes the form of standard CAPM with the commodity
index being used to proxy the return on the market. Similar to Draper (1984), we estimate the
market model as
8
ctmtccct RR
ε
β
α
+
+
=
where Rmt is the return on the Dow Jones AIG Commodity Index on day t. Dow Jones AIG
commodity index is a highly liquid and diversified benchmark for the commodities market
consisting of nineteen physical commodities and is used in the model as a proxy for the
commodities market performance. Assuming the error term, ct
ε
, has an expected value of zero,
the error term is in fact the abnormal return at t in event time in the vicinity of event c denoted
as
ctct
AR
ε
=
1.
2. The ARCH Model: Similar to Deaves and Krinsky (1992), we estimate an ARCH model of oil
returns
ctcct eR
+
=
µ
where the variance of the error term is defined as
=
+= q
iictjccctct ee
1
2
0
2)(
δδσ
with the model estimated up to seven lags, i.e. q=7. This model provides a second estimate for
the abnormal return at t in event time in the vicinity of event c denoted as
ctct eAR
=
2.
3. The Fama-French Model: The third model we use to measure normal performance is the
multi-factor model suggested by Fama and French (1993). They suggest a three factor model
utilizing the size (SMB), book to market ratio (HML) and the risk premium on the market
portfolio as determinants of asset returns. The return model takes the form
cttctcftmtccct aHMLbSMBbRRbbR
+
+
+
+= 3210 )(
Based on this model, we get a third estimate for abnormal returns denoted as
ctct aAR
=
3.
9
3.2. Model Estimation and Hypothesis Testing
Assuming t=0 is the event date, the model for normal performance is estimated over the estimation
period -80 days to -21 days covering about two months prior to the event. Once the parameters for the
normal performance models are estimated and announcements (or events) categorized (see Table
abnormal returns are then calculated using equation 1 over the twenty business days prior to and
subsequent to the end of each event (announcement), i.e. -20 days to +20 days. In some cases, due to
the proximity of conferences (in particular, in the case of extraordinary meetings), several events had
to be dropped in order to prevent the event from influencing the normal performance model par
estimates. Onc
2),
ameter
e daily abnormal returns are calculated, average daily abnormal returns are then
calculated as
=
n
e
sum of
to a
=
cct
tnARAR
1
/ (2)
where n is the number of events for which observations exist for a given type of event. The averag
abnormal returns are finally used to calculate cumulative average abnormal returns as the
average abnormal returns for day -20 the specified day T using the following formul
=
=
t
T
ce
t
including 5, 10, 15 and 20 days subsequent to the end of the
event. Deaves and Krinsky (1992) suggest that under the null hypothesis of market efficiency, the
expected return can be formulated as
T
tARCAR
20
(3)
In order to see whether the market reacted efficiently to these announcements, we test the significan
of cumulative average abnormal returns following the announcement date. More specifically, we tes
for possible persistence in the CAR paths following the announcement, i.e., the null hypothesis H0:
CART-CAR1 = 0 for selected T values
==
TT
t
n
cctT
21
1
=
=∑∑ =t
t
AREnARECARCARE
2
/)(.
10
Assuming the error terms (i.e. the abnormal returns) are i.i.d. with an expected return of zero and
constant variance of σ2, the variance term is then formulated as
nTARCARCAR T
t
t
T/)1(var)var( 2
2
1
σ
=
=
=
.
4. Empirical Findings
4.1. OPEC Announcements
ries are
the
he paths
ive
roduction cut. We observe increasingly
positive CAR values during the post announcement period beyond t=1. One explanation for this
production level, i.e. no change. For all the normal performance models utilized, we observe gradually
The paths of the cumulative abnormal returns for each of the three OPEC announcement catego
illustrated in Figures 1, 2, and 3 plotting the CAR’s from the market model, the ARCH model and
Fama-French model, respectively. Within each OPEC announcement category, separate plots are
reported for spot and the three futures return series representing different maturities. First, it is
apparent that the largest reaction to the announcement is observed in the case of spot and shorter
maturity futures contracts. Consistent across the three different normal return models used to estimate
abnormal returns, we find larger abnormal returns for spot and shorter maturity futures contracts.
Second, the OPEC production increase announcements seem to have no significant effect on t
of cumulative abnormal returns. However, consisntent with Guidi, Russel and Tarbert (2006), we find
a significant reaction to OPEC production cut announcements with significantly positive cumulat
abnormal returns following the announcement. Furthermore, there seems to exist a degree of
persistence in CAR paths following the announcement of a p
observation might be that the market shows an incomplete reaction to the announcement leading to
further price adjustments during post announcement period.
Interestingly, the market seems to react to OPEC’s announcements on maintaining the current
11
decreasing CARs following the announcement, suggesting downward price adjustments after OP
announces no changes i
EC
n production levels. Similar to production cut announcements, we observe an
incomplete reaction to the announcement leading to persistence in CAR patterns during the post
announcement period.
l
et, more
2
l the
announcement date (t=0) suggesting some anticipation of the outcome by the market.This is not
surprising as information leakages are possible, even before the conference begins.
t
n
ormal
sitive
Another observation is that, with the exception of abnormal returns estimated using the market mode
(Figure 1), the information content of the final release seems to be anticipated by the mark
apparently in the case of production cut and no change announcements. As can be seen in Figures
and 3, CARs prior to production cut announcements seem to gradually increase unti
After this visual inspection of CAR paths, we formally test whether the market reacts efficiently to
these announcements. For this purpose, we test the significance of the difference between cumulative
average residuals for the day following the announcement and those at a future date during the pos
event period. Table 3 reports the test results for the null hypothesis H0: CART-CAR1 = 0 for selected T
values including 5, 10, 15 and 20 days subsequent to the announcement. Under the null of market
efficiency, one would expect this difference to have an expected value of zero as the market’s reactio
to the announcement would be completed at t=1. Therefore, a significant value for (CART-CAR1)
would indicate an incomplete reaction to the announcement which leads to persistence in abn
returns during the post event period. Consistent with our visual inspection of Figures 1-3, we find no
evidence to reject the null hypothesis of efficiency in the case of OPEC production increase
announcements. However, in the case of production cut announcements, we find significant po
returns during the post event period indicating incomplete reaction of investors to the announcement.
Consider the case of T=20 where the market has about three weeks to digest the information.
Regardless of the normal performance model used to estimate abnormal returns, all CARs are found to
12
be statistically significant and positive. Positive returns estimated take on values as high as 7.58%,
7.17%, 6.07%, 4.60% for the spot market, nearest futures contract, third closest contract, and twelfth
closest contract respectively. Considering the fact that these returns (CAR20-CAR1) are approxim
one month returns, they amount to annual returns of as high as 90% for the spot market and 55% fo
the longest maturity futures market. Clearly, these profit opportunities are not only statistically
significant but also economically significant too. As a result, transaction costs cannot explain this
abnormal performance as futures markets (in particular energy futures) have evolved to be one of the
most liquid markets. Similarly, one can suggest greater return volatility during the event period to
explain the abnormal performance. However, in a study of implied volatilities from options on cr
oil futures surrounding OPEC meetings, Horan, Peterson, and Mahar (2004) find that highly visible bi-
annual conferences are associated with a drop, not increase, in volatility. Furthermore, the most
pronounced decline in volatility coincides with the meetings of
ately
r
ude
the Ministerial Monitoring Committee,
which makes production recommendations to the larger conference. Therefore, none of these two
arguments can be used to explain this abnormal performance.
d. In short, consistent with Guidi,
Russel and Tarbert (2006), we find a significant market reaction to production cut announcements with
substantial abnormal returns during the post announcement period.
4.2. Strategic Petroleum Reserve (SPR) Announcements
sed from
ama-
Regarding OPEC announcements to maintain current production levels, i.e. no change, we find some
evidence of return persistence, but only on the downside. However, the statistical tests do not lead to
consistent results across the three different normal return models use
Figures 4-6 illustrate CAR paths around SPR announcements. Once again, each SPR press release is
one event and each event is classified into two categories: SPR increase (when oil is purcha
the market) and SPR decrease (when oil is released to the market). With the exception of the F
13
French model based abnormal returns (in Figure 6), we find that the market reacts to these
announcements with downward (upward) price adjustments after SPR drawdown (purchase
announcements. Visual inspection of CAR paths in Figures 4 and 5 suggests that the government’s us
of this program initiates a short-run market reaction lasting for about a week following the
announcement date. However, statistical tests of cumulative ab
)
e
normal return differences reported in
Table 4, do not yield consistently significant differences for (CART-CAR1) suggesting no statistically
significant impact of these announcements on the oil market.
5. Conclusions
he
petroleum supply interruption on economic
activity by having sufficient petroleum reserves. Therefore, this study sheds a light on the impact of the
government’s use of the reserves on oil market activity.
se
ase
This paper uses the event study methodology to study the effects of OPEC conferences and U.S.
Strategic Petroleum Reserve announcements on the crude oil market activity in the U.S. We examine a
total of 63 OPEC press releases and 15 SPR announcements over the period 1983-2008 and test
whether the market reaction to these announcements conforms to the efficient market hypothesis. T
study has several contributions. First, we provide evidence from both spot and futures markets in order
to examine whether these events affect more short or long end of the crude oil market. Second, we
perform separate tests by using three different measures of abnormal (excess) returns estimated using
the market model, the ARCH model, and the three-factor Fama-French model. Third, we include in
our analysis the events regarding the Department of Energy’s SPR announcements. The SPR Program
was established to prevent the negative effects of a major
Our tests of cumulative abnormal returns suggest no significant reaction to OPEC production increa
announcements regardless of the normal peformance model used in the analysis. However, in the c
of production cut announcements, we find significant positive returns during the post event period
14
indicating incomplete reaction of investors to the announcement. The degree of return persistence
following OPEC production cut announcements creates substantial excess returns to investors who
take long positions on the day following the end of OPEC conferences. We find that these excess
returns can amount to annual returns of as high as 90% in the spot market and 55% in the longest
maturity fu
tures market. Similarly, in the case of OPEC announcements to maintain current production
levels, i.e. no change in production levels, we find some evidence of return persistence, however on the
downside.
tes a short-
rn differences suggest no statistically significant impact of these
announcements on the oil market, suggesting that the SPR Program may not be an effective tool to
stabilize the oil market.
In the case of SPR announcements, we find that the government’s use of this program initia
run market reaction lasting for about a week following the announcement date. However, tests of
cumulative abnormal retu
15
References
Binder, J. J., 1998, The Event Study Methodology Since 1969, Review of Quantitative Finance and
Accounting, 11 (1998): 111–137.
Brown, S.J. and J.B. Warner, "Measuring Security Price Performance," Journal of Financial
Economics, 8 (September 1980): 205-258.
Brown, S.J. and J.B. Warner, "Using Daily Stock Returns: The Case of Event Studies," Journal of
Financial Economics, 14 (March 1985): 3-31.
Considine, Timothy J., 2006. Is the Strategic Petroleum Reserve our Ace in the Hole? Energy Journal
Vol. 27, Issue 3, pp. 91-112.
Deaves, Richard and Itzhak Krinsky, 1992. The Behaviour of Oil Futures Returns around OPEC
Conferences. Journal of Futures Markets, Vol. 12, No. 5. pp. 563-574.
Draper , Dennis, 1984. The Behavior of Even-Related Returns on Oil Futures Contracts. Journal of
Futures Markets, Vol. 4, No. 2. pp. 125-132.
Fama, E.F., French, K.R., 1993." Common Risk Factors in the Returns on Stocks and Bonds," Journal
of Financial Economics, 33 (1993): 3 – 56.
Guidi, Marco G.D., Alexander Russell, and Haether Tarbert, 2006. The Effects of OPEC Policy
decisions on Oil and Stock Prices. Organization of the Petroleum Exporting Countries, March 2006,
pp. 1-18.
Horan, Stephen M., Jeffrey H. Peterson, & James Mahar, 2004. Implied Volatility of Oil Futures
Options Surrounding OPEC Meetings. The Energy Journal, 25 (3), 103-125.
MacKinlay, A. Craig, 1997. "Event Studies in Economics and Finance", Journal of Economic
Literature, March 1997. (added 1/98)
Peterson, Pamela P., 1989. "Event Studies: A Review of Issues and Methodology", Quarterly Journal
of Business and Economics, 28, pages 36-66, 1989. (added 8/99)
Taylor, Jerry and Peter Van Doren, 2005. The Case against the Strategic Petroleum Reserve. Policy
Analysis No. 555, pp. 1-21.
Thompson, R., 1995. Empirical methods of event studies in coporate finance in Jarrow, R. A.,
Maksimovic, V. and Ziemba, W. T. (eds.), Handbook in Operations Research and Management
Science-Finance, Elsevier.
Wirl, F. & A. Kujundzic, 2004. The Impact of OPEC Conference Outcomes on World Oil Prices
19842001. The Energy Journal, 25 (1), 4561
16
Table 1: Descriptive Statistics
Sample Period (March 30, 1983 June 1, 2008)
N = 6303
Spot Future1 Future3 Future12
Mean 0.022% 0.044% 0.041% 0.051%
Std. Dev. 2.376% 2.056% 1.788% 1.389%
Maximum -40.204% -38.407% -28.427% -12.948%
Minimum 19.861% 12.353% 11.256% 7.976%
Skewness -1.014 -1.260 -0.850 -0.424
Kurtosis 19.375 23.261 13.643 4.501
ρs,f 0.850 0.813 0.729
Note: ρs,f is the correlation coefficient between spot and futures return series.
Table 2: OPEC and SPR Announcements.
Announcements
OPEC production hike 17
OPEC production cut 21
OPEC no change 25
Total OPEC 63
SPR decrease (Release to the market) 11
SPR increase (Purchase from the market) 4
Total SPR 15
Note: Each OPEC conference or SPR press release is one event. ‘SPR decrease’
indicates announcements on the release of crude oil to the market from the Strategic
Petroleum Reserves. ‘SPR increase’ indicates announcements on crude oil purchases
to build SPR inventories.
17
Table 3: CAR Differences (%) and Absolute t-Statistics (OPEC Announcements).
Event Day of CAR
Increment Relative
to CAR for t=1
Spot Returns Nearest Contract, F1 Third-closest Contract, F3 Twelfth-closest Contract, F12
CAR1 CAR2 CAR3 CAR1 CAR2 CAR3 CAR1 CAR2 CAR3 CAR1 CAR2 CAR3
OPEC Increase
5 0.42
rease
ang
0.60 0.27 0.23 0.25 0.32 0.29 0.42 0.65 0.48 0.24 0.18
(0.40) (0.46) (0.20) (0.23) (0.21) (0.26) (0.31) (0.39) (0.55) (0.57) (0.26) (0.19)
10 0.40 -0.79 -1.83 -0.23 -1.18 -2.09 0.18 -0.87 -1.69 0.84 -0.33 -0.89
(0.25) (0.41) (0.90) (0.15) (0.68) (1.12) (0.20) (0.26) (0.37) (0.38) (0.17) (0.13)
15 0.20 0.52 -1.14 -0.52 -0.09 -1.38 -0.28 0.07 -0.90 0.15 0.17 -0.95
(0.10) (0.21) (0.45) (0.12) (0.55) (0.90) (0.16) (0.21) (0.30) (0.30) (0.14) (0.10)
20 -1.18 -1.87 -3.22 -1.39 -1.80 -3.16 -0.94 -1.19 -2.23 -0.06 -0.72 -1.84
(0.51) (0.66) (1.09) (0.10) (0.47) (0.77) (0.14) (0.18)
(0.25)
(0.26)
(0.12)
(0.09)
OPEC Dec
5 2.06*2.90** 1.56 1.45*2.60** 1.21 1.17*2.09** 0.99*0.70*2.09** 1.32*
(1.74) (2.21) (1.12) (1.87) (2.30) (1.07) (1.79) (2.25) (1.07) (1.26) (2.63) (1.60)
10 1.83 3.83** 2.35 1.47 3.25** 1.66 1.34*2.60** 1.59 0.51 2.01*1.42
(1.03) (1.95) (1.12) (1.26) (1.91) (0.97) (1.38) (1.86) (1.14) (0.62) (1.68) (1.15)
15 3.77*6.91*** 4.93** 3.01** 6.26*** 3.72** 3.24*** 5.51*** 3.60** 2.04** 4.35*** 3.28**
(1.71) (2.82) (1.89) (2.07) (2.95) (1.75) (2.66) (3.17) (2.07) (1.96) (2.91) (2.13)
20 3.41*7.58*** 4.15*2.90** 7.17*** 3.21 3.08** 6.07*** 3.28** 1.61*4.60*** 2.80**
(1.32)
e
(2.65) (1.37) (1.71) (2.90)
(1.30)
(2.17)
(3.00)
(1.62)
(1.33)
(2.65)
(1.56)
OPEC No Ch
5 -0.76 -0.59 -0.53 -1.53** -1.22 -1.31 -1.43** -1.09 -1.23 -1.01** -0.79 -1.01
(1.01) (0.67) (0.52) (2.11) (1.26) (1.15) (2.32) (1.37) (1.30) (2.15) (1.17) (1.34)
10 -1.55*-2.37*-2.48*-2.10** -2.43*-2.63*-2.24** -2.33** -2.58** -1.57** -1.82*-2.11**
(1.38) (1.81) (1.62) (1.93) (1.68) (1.53) (2.42) (1.95) (1.82) (2.22) (1.80) (1.86)
15 -2.15*-2.13 -3.09*-2.15*-1.68 -2.52 -2.34** -1.72 -2.52*-1.38*-0.86 -1.69
(1.53) (1.30) (1.62) (1.58) (0.93) (1.18) (2.03) (1.15) (1.43) (1.56) (0.68) (1.19)
20 -2.95** -2.53*-4.27** -3.10** -2.15 -3.78*-3.15** -2.17 -3.64** -1.53*-1.02 -2.19*
(1.80) (1.33) (1.92) (1.96) (1.02) (1.52) (2.35) (1.25) (1.77) (1.49) (0.70) (1.33)
Note: The significance of the difference between cumulative abnormal returns for day T and day 1 (for T=5, 10, 15, and 20) are tested. CAR1, CAR2, and CAR3
represent the abnormal returns obtained using the market model, ARCH model and the Fama-French model, respectively. Differences marked with a *** (**, *) are
statistically significant at the 1% (5%, 10%) level. Absolute t-statistics are in brackets.
Table 4: CAR Differences (%) and Absolute t-Statistics (SPR Announcements).
Event Day of CAR
Increment Relative
to CAR for t=1
Spot Returns Nearest Contract, F1 Third-closest Contract, F3 Twelfth-closest Contract, F12
CAR1 CAR2 CAR3 CAR1 CAR2 CAR3 CAR1 CAR2 CAR3 CAR1 CAR2 CAR3
SPR Increase
5
ease
0.30 2.96 3.54*0.99 3.34*3.95*0.68 2.66 3.01*0.11 1.56 1.76
(0.19) (1.38) (1.58) (0.72) (1.63) (1.89) (0.55) (1.44) (1.61) (0.10) (1.05) (1.16)
10 -0.06 4.31 3.67 1.06 4.73*4.63*0.84 3.97*3.58 0.63 2.88 2.90
(0.02) (1.34) (1.09) (0.51) (1.54) (1.48) (0.45) (1.44) (1.28) (0.39) (1.29) (1.28)
15 -0.09 4.73 4.32 1.47 6.03*6.56*1.50 5.54*5.63*1.19 4.15*4.95**
(0.03) (1.18) (1.03) (0.57) (1.57) (1.68) (0.65) (1.61) (1.62) (0.59) (1.49) (1.75)
20 -1.58 3.85 2.59 0.61 5.46 6.18*1.18 5.50*5.80*1.23 4.42*5.88**
(0.46) (0.82) (0.53) (0.20) (1.22) (1.36) (0.44) (1.37)
(1.43)
(0.53)
(1.36)
(1.78)
SPR Decr
5 -2.20*3.54 4.87*-2.20*1.54 3.49 -1.42 0.76 2.44 -0.68 0.48 1.02
(1.44) (1.04) (1.58) (1.76) (0.46) (1.14) (1.36) (0.32) (1.17) (0.84) (0.30) (0.69)
10 -2.30 1.32 5.17 -2.10 2.08 7.03*-1.12 1.48 5.84** -0.32 1.52 3.04*
(1.01) (0.26)
(1.12) (1.12) (0.41) (1.52) (0.71) (0.42) (1.86) (0.26) (0.63) (1.37)
15 0.42 2.50 8.48*-1.51 2.28 9.17*-1.16 1.21 7.00** -0.45 1.63 3.31
(0.15) (0.39)
(1.47)
(0.65) (0.36) (1.59) (0.59) (0.28) (1.79) (0.29) (0.55) (1.20)
20 -3.97 -0.51 6.77 -3.93*1.35 9.35*-3.35*0.20 6.87*2.07 1.03 2.91
(1.20) (0.07) (1.01) (1.44) (0.18) (1.39) (1.47) (0.04) (1.51) (1.16) (0.30) (0.90)
Note: The significance of the difference between cumulative abnormal returns for day T and day 1 (for T=5, 10, 15, and 20) are tested. CAR1, CAR2, and CAR3
represent the abnormal returns obtained using the market model, ARCH model and the Fama-French model, respectively. Differences marked with a *** (**, *) are
statistically significant at the 1% (5%, 10%) level. Absolute t-statistics are in brackets.
19
Figure 1. Cumulative abnormal returns for OPEC announcements from event day -20 to event day 20.
The abnormal return is calculated using the market model as the normal return measure.
Spot Returns
-6.000%
-5.000%
-4.000%
-3.000%
-2.000%
-1.000%
0.000%
1.000%
2.000%
3.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
Increase No Change Decrease
Clo ses t Future s Contra ct
-5.000%
-4.000%
-3.000%
-2.000%
-1.000%
0.000%
1.000%
2.000%
-20-18-16-14-12-10-8-6-4-202468101214161820
Eve n t T im e
CAR
Increase No Change Decrease
Third Close st Futures Contract
-5.000%
-4.000%
-3.000%
-2.000%
-1.000%
0.000%
1.000%
2.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
Increase No Change Decrease
Twelfth Closest Futures Contract
-4.000%
-3.000%
-2.000%
-1.000%
0.000%
1.000%
2.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
Increase No Change Decrease
Figure 2. Cumulative abnormal returns for OPEC announcements from event day -20 to event day 20.
The abnormal return is calculated using the ARCH model as the normal return measure.
Spot Returns
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
8.000%
10.000%
-20-18-16-14-12-10-8-6-4-202468101214161820
Eve n t T im e
CAR
Increase No Change Decrease
Close st Futures Contract
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
8.000%
10.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
Increase No Change Decrease
Third Close st Futures Contract
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
8.000%
10.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
Increase No Change Decrease
Twelfth Closest Futures Contract
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
8.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
Increase No Change Decrease
21
Figure 3. Cumulative abnormal returns for OPEC announcements from event day -20 to event day 20.
The abnormal return is calculated using the Fama-French model as the normal return measure.
Spot Returns
-8.000%
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
Increase No Change Decrease
Close st Futures Contract
-8.000%
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
Increase No Change Decrease
Sixth Closest Futures Contract
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
Increase No Change Decrease
Twelfth Closest Futures Contract
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
Increase No Change Decrease
22
Figure 4. Cumulative abnormal returns for SPR announcements from event day -20 to event day 20.
The abnormal return is calculated using the market model as the normal return measure.
Spot Returns
-10.000%
-8.000%
-6.000%
-4.000%
-2.000%
0.000%
2.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t Ti m e
CAR
SPR Increase S PR Decr eas e
Close st Futures Contract
-8.000%
-6.000%
-4.000%
-2.000%
0.000%
2.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
SPR Increase SPR Dec re ase
Third Clo sest Future s Contrac t
-8.000%
-6.000%
-4.000%
-2.000%
0.000%
2.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
SPR Increase SPR Dec re ase
Twelfth Closest Futures Contract
-6.000%
-5.000%
-4.000%
-3.000%
-2.000%
-1.000%
0.000%
1.000%
2.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
SPR Increase SPR Decr eas e
23
Figure 5. Cumulative abnormal returns for SPR announcements from event day -20 to event day 20.
The abnormal return is calculated using the ARCH model as the normal return measure.
Spot Returns
-10.000%
-8.000%
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
SPR Increase SPR Dec rea se
Closes t Futures Contract
-8.000%
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
8.000%
10.000%
-20-18-16-14-12-10-8-6-4-202468101214161820
Eve n t T im e
CAR
SPR Increase SPR Dec rea se
Third Closest Future s Contract
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
8.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
SPR Increase SPR Decr ease
Twelfth Closest Futures Contract
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
SPR Increas e SPR Decr eas e
24
Figure 6. Cumulative abnormal returns for SPR announcements from event day -20 to event day 20.
The abnormal return is calculated using the Fama-French model as the normal return measure.
Spot Returns
-10.000%
-8.000%
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
8.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
SPR Increase SPR Decrease
Closes t Futures Contract
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
8.000%
10.000%
12.000%
-20-18-16-14-12-10-8-6-4-202468101214161820
Eve n t T im e
CAR
SPR Increase SPR Dec rea se
Third Closest Future s Contract
-6.000%
-4.000%
-2.000%
0.000%
2.000%
4.000%
6.000%
8.000%
10.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
SPR Increase SPR Dec rea se
Twelfth Closest Futures Contract
-4.000%
-2.000%
0.000%
2.000%
4.000%
-20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20
Eve n t T im e
CAR
SPR Increas e SPR Decr eas e
25
APPENDIX: Abnormal and cumulative abnormal returns during the event period.
Table 5. Abnormal and Cumulative Abnormal Returns Around OPEC Announcements. (The
abnormal return is calculated using the market model as the normal return measure)
Market Model (Spot Returns)
Event OPEC Increase OPEC Decrease OPEC No Change
Day AR CAR AR CAR AR CAR
-20 0.326% 0.326% 0.223% 0.223% -0.332% -0.332%
-19 -0.177% 0.149% -0.353% -0.130% -0.048% -0.380%
-18 0.811% 0.960% 0.092% -0.038% 0.070% -0.310%
-17 -0.895% 0.065% -0.261% -0.298% -0.028% -0.337%
-16 0.890% 0.955% -0.321% -0.619% 0.223% -0.114%
-15 0.049% 1.005% -1.174% -1.794% 0.102% -0.012%
-14 -0.135% 0.870% -0.031% -1.825% -0.455% -0.467%
-13 -0.328% 0.542% 0.006% -1.819% 0.434% -0.033%
-12 0.019% 0.561% -0.790% -2.608% 0.434% 0.400%
-11 -0.003% 0.558% 0.004% -2.604% 0.102% 0.502%
-10 0.885% 1.442% -0.282% -2.886% 0.520% 1.022%
-9 -0.760% 0.683% -0.252% -3.138% 0.168% 1.190%
-8 0.416% 1.099% -0.489% -3.627% -0.018% 1.172%
-7 0.652% 1.751% -0.262% -3.889% -0.028% 1.144%
-6 -0.126% 1.625% -0.399% -4.288% -0.649% 0.495%
-5 -0.916% 0.709% -0.643% -4.931% -0.168% 0.327%
-4 -0.821% -0.112% 0.142% -4.789% -0.220% 0.107%
-3 0.823% 0.710% -0.077% -4.866% -1.103% -0.996%
-2 1.096% 1.807% 0.486% -4.381% 0.151% -0.845%
-1 -0.785% 1.021% 0.393% -3.987% -0.180% -1.025%
0 -0.562% 0.460% -0.297% -4.284% -0.154% -1.179%
1 -0.316% 0.144% -0.172% -4.456% 0.605% -0.575%
2 0.550% 0.694% 0.254% -4.202% 0.005% -0.569%
3 -0.323% 0.371% 0.346% -3.856% 0.209% -0.360%
4 -0.061% 0.310% 0.900% -2.956% -0.661% -1.022%
5 0.254% 0.565% 0.557% -2.399% -0.308% -1.330%
6 -0.054% 0.510% 0.976% -1.423% -0.534% -1.864%
7 0.153% 0.663% -0.252% -1.675% -0.156% -2.020%
8 -0.692% -0.029% 0.277% -1.398% -0.070% -2.090%
9 -0.089% -0.118% -0.224% -1.622% -0.005% -2.096%
10 0.658% 0.540% -1.001% -2.623% -0.031% -2.126%
11 -0.389% 0.151% 0.811% -1.811% -0.418% -2.544%
12 0.071% 0.222% 0.242% -1.569% 0.453% -2.091%
13 -0.059% 0.163% 0.026% -1.543% -0.344% -2.435%
14 -0.172% -0.009% 0.108% -1.435% 0.144% -2.291%
15 0.351% 0.342% 0.751% -0.684% -0.429% -2.720%
16 0.128% 0.470% -1.471% -2.155% -0.609% -3.330%
17 -0.307% 0.163% -0.304% -2.459% 0.362% -2.967%
18 -0.567% -0.404% 1.651% -0.808% -0.721% -3.688%
19 -0.313% -0.717% 0.295% -0.513% -0.116% -3.804%
20 -0.318% -1.036% -0.529% -1.042% 0.283% -3.521%
Note: AR is the sample average abnormal return for the specified day in event time and CAR is the
sample average cumulative abnormal return for day -20 to the specified day. Event time is days
relative to the announcement date.
26
(Table 5 continued)
Market Model (Nearby Futures Contract, F1)
Event OPEC Increase OPEC Decrease OPEC No Change
Day AR CAR AR CAR AR CAR
-20 0.308% 0.308% 0.085% 0.085% -0.198% -0.198%
-19 -0.044% 0.264% -0.314% -0.229% -0.108% -0.306%
-18 0.509% 0.773% -0.327% -0.555% 0.174% -0.132%
-17 -0.898% -0.126% 0.245% -0.310% 0.204% 0.072%
-16 0.482% 0.357% -0.575% -0.886% 0.230% 0.303%
-15 -0.077% 0.279% -1.012% -1.897% -0.185% 0.117%
-14 -0.175% 0.104% 0.145% -1.752% -0.074% 0.043%
-13 -0.318% -0.214% -0.228% -1.980% 0.275% 0.318%
-12 -0.080% -0.294% -0.599% -2.579% 0.077% 0.395%
-11 -0.381% -0.675% -0.191% -2.770% 0.377% 0.771%
-10 0.901% 0.226% -0.368% -3.139% 0.540% 1.312%
-9 -0.752% -0.527% -0.121% -3.260% -0.015% 1.296%
-8 0.200% -0.327% -0.213% -3.472% -0.169% 1.128%
-7 0.565% 0.238% -0.589% -4.061% -0.260% 0.868%
-6 -0.052% 0.186% 0.226% -3.836% -0.375% 0.493%
-5 -0.614% -0.428% -0.453% -4.288% -0.375% 0.118%
-4 -0.295% -0.723% 0.396% -3.892% -0.057% 0.062%
-3 0.165% -0.559% -0.234% -4.126% -1.022% -0.960%
-2 1.469% 0.910% -0.018% -4.144% 0.253% -0.708%
-1 -1.205% -0.295% 0.183% -3.962% -0.216% -0.923%
0 0.117% -0.178% -0.279% -4.241% -0.210% -1.134%
1 -0.254% -0.432% -0.004% -4.245% 0.758% -0.375%
2 0.302% -0.130% 0.337% -3.908% -0.382% -0.757%
3 -0.551% -0.681% 0.348% -3.560% 0.073% -0.684%
4 -0.017% -0.698% 0.350% -3.210% -0.534% -1.219%
5 0.500% -0.198% 0.415% -2.795% -0.688% -1.906%
6 0.017% -0.181% 0.698% -2.097% -0.344% -2.250%
7 -0.664% -0.845% -0.045% -2.141% -0.085% -2.335%
8 -0.247% -1.092% -0.112% -2.253% 0.047% -2.288%
9 -0.242% -1.334% -0.222% -2.475% -0.014% -2.302%
10 0.676% -0.658% -0.304% -2.780% -0.178% -2.480%
11 -0.070% -0.728% 0.737% -2.043% -0.100% -2.579%
12 0.223% -0.505% 0.107% -1.937% 0.378% -2.201%
13 -0.359% -0.863% -0.115% -2.051% -0.179% -2.380%
14 -0.219% -1.082% 0.314% -1.737% 0.246% -2.133%
15 0.133% -0.949% 0.504% -1.233% -0.388% -2.521%
16 -0.183% -1.132% -0.441% -1.674% -0.751% -3.272%
17 0.332% -0.801% -0.407% -2.081% 0.230% -3.042%
18 -0.051% -0.852% 0.554% -1.527% -0.614% -3.656%
19 -1.080% -1.932% 0.302% -1.225% -0.231% -3.887%
20 0.108% -1.824% -0.123% -1.348% 0.410% -3.476%
27
(Table 5 continued)
Market Model (Third Closest Contract, F3)
Event OPEC Increase OPEC Decrease OPEC No Change
Day AR CAR AR CAR AR CAR
-20 0.158% 0.158% -0.060% -0.060% -0.184% -0.184%
-19 -0.128% 0.030% -0.225% -0.285% -0.040% -0.223%
-18 0.475% 0.505% -0.256% -0.541% 0.092% -0.131%
-17 -0.725% -0.220% 0.277% -0.265% 0.076% -0.055%
-16 0.380% 0.160% -0.480% -0.745% 0.203% 0.148%
-15 0.049% 0.209% -0.789% -1.534% -0.212% -0.063%
-14 -0.217% -0.007% 0.048% -1.485% -0.161% -0.225%
-13 -0.221% -0.228% -0.091% -1.576% 0.207% -0.018%
-12 -0.213% -0.441% -0.412% -1.989% 0.060% 0.042%
-11 -0.460% -0.901% -0.104% -2.092% 0.291% 0.333%
-10 0.816% -0.085% -0.240% -2.332% 0.396% 0.728%
-9 -0.653% -0.738% 0.142% -2.191% -0.136% 0.592%
-8 0.174% -0.564% -0.144% -2.335% -0.153% 0.439%
-7 0.574% 0.010% -0.378% -2.713% -0.255% 0.184%
-6 0.069% 0.079% 0.168% -2.546% -0.231% -0.047%
-5 -0.493% -0.414% -0.314% -2.860% -0.392% -0.439%
-4 -0.142% -0.556% 0.355% -2.505% 0.012% -0.428%
-3 0.268% -0.287% -0.195% -2.700% -0.815% -1.243%
-2 1.334% 1.047% 0.035% -2.664% 0.082% -1.161%
-1 -1.318% -0.271% 0.002% -2.662% -0.304% -1.465%
0 0.049% -0.222% -0.224% -2.886% -0.216% -1.681%
1 -0.029% -0.251% 0.134% -2.752% 0.692% -0.989%
2 0.269% 0.018% 0.213% -2.538% -0.379% -1.367%
3 -0.371% -0.352% 0.269% -2.269% 0.073% -1.294%
4 -0.004% -0.356% 0.229% -2.040% -0.490% -1.785%
5 0.396% 0.040% 0.454% -1.586% -0.630% -2.415%
6 -0.020% 0.020% 0.538% -1.047% -0.253% -2.668%
7 -0.600% -0.580% -0.003% -1.051% -0.178% -2.846%
8 -0.137% -0.716% 0.013% -1.037% 0.043% -2.803%
9 -0.104% -0.820% -0.188% -1.225% -0.062% -2.865%
10 0.753% -0.067% -0.182% -1.408% -0.360% -3.226%
11 -0.131% -0.198% 0.652% -0.756% -0.205% -3.431%
12 0.235% 0.037% 0.217% -0.539% 0.329% -3.102%
13 -0.240% -0.203% -0.057% -0.597% -0.155% -3.257%
14 -0.217% -0.421% 0.476% -0.121% 0.208% -3.049%
15 -0.107% -0.528% 0.605% 0.485% -0.275% -3.324%
16 -0.220% -0.748% -0.417% 0.067% -0.583% -3.907%
17 0.314% -0.433% -0.401% -0.334% 0.172% -3.734%
18 -0.007% -0.441% 0.460% 0.127% -0.599% -4.334%
19 -0.890% -1.330% 0.212% 0.338% -0.126% -4.459%
20 0.139% -1.191% -0.011% 0.328% 0.320% -4.139%
28
(Table 5 continued)
Market Model (Twelfth Closest Contract, F12)
Event OPEC Increase OPEC Decrease OPEC No Change
Day AR CAR AR CAR AR CAR
-20 0.030% 0.030% -0.119% -0.119% -0.167% -0.167%
-19 -0.146% -0.117% -0.014% -0.134% 0.137% -0.030%
-18 0.323% 0.206% -0.101% -0.234% -0.013% -0.042%
-17 -0.253% -0.047% 0.373% 0.139% -0.098% -0.140%
-16 0.375% 0.328% 0.004% 0.143% 0.082% -0.058%
-15 0.126% 0.455% -0.374% -0.232% -0.176% -0.234%
-14 -0.376% 0.079% 0.031% -0.201% -0.210% -0.444%
-13 -0.130% -0.051% -0.004% -0.205% 0.105% -0.339%
-12 -0.501% -0.552% -0.282% -0.488% 0.095% -0.244%
-11 -0.352% -0.904% -0.150% -0.638% 0.176% -0.067%
-10 0.783% -0.121% -0.278% -0.915% 0.164% 0.096%
-9 -0.451% -0.572% 0.309% -0.606% -0.246% -0.149%
-8 0.191% -0.381% 0.033% -0.573% -0.075% -0.225%
-7 0.497% 0.117% -0.368% -0.941% -0.212% -0.436%
-6 0.401% 0.518% 0.232% -0.709% -0.025% -0.461%
-5 -0.287% 0.232% -0.265% -0.974% -0.337% -0.799%
-4 -0.083% 0.148% 0.194% -0.780% 0.242% -0.556%
-3 0.215% 0.363% -0.104% -0.884% -0.549% -1.105%
-2 1.084% 1.448% 0.113% -0.771% 0.183% -0.922%
-1 -1.218% 0.230% 0.124% -0.648% -0.386% -1.308%
0 0.181% 0.411% -0.259% -0.907% -0.124% -1.432%
1 0.254% 0.665% 0.068% -0.839% 0.399% -1.033%
2 0.258% 0.923% 0.122% -0.718% -0.453% -1.485%
3 -0.292% 0.632% 0.183% -0.535% 0.087% -1.398%
4 0.276% 0.908% -0.132% -0.667% -0.189% -1.588%
5 0.239% 1.147% 0.525% -0.142% -0.459% -2.047%
6 0.061% 1.208% 0.219% 0.077% -0.109% -2.156%
7 -0.565% 0.643% -0.211% -0.133% -0.205% -2.360%
8 0.014% 0.657% 0.105% -0.028% 0.075% -2.286%
9 0.166% 0.823% -0.162% -0.191% 0.087% -2.198%
10 0.685% 1.508% -0.136% -0.327% -0.407% -2.605%
11 -0.139% 1.369% 0.465% 0.138% -0.171% -2.776%
12 0.010% 1.379% 0.223% 0.360% 0.299% -2.476%
13 -0.108% 1.270% -0.268% 0.092% -0.024% -2.501%
14 -0.099% 1.172% 0.555% 0.647% 0.255% -2.245%
15 -0.358% 0.814% 0.550% 1.197% -0.163% -2.408%
16 -0.239% 0.575% -0.370% 0.827% -0.390% -2.799%
17 0.349% 0.924% -0.612% 0.216% 0.239% -2.559%
18 0.142% 1.066% 0.368% 0.584% -0.266% -2.826%
19 -0.748% 0.318% 0.090% 0.674% 0.062% -2.763%
20 0.285% 0.602% 0.095% 0.769% 0.200% -2.563%
29
Table 6. Abnormal and Cumulative Abnormal Returns Around OPEC Announcements. (The abnormal
return is calculated using the ARCH model as the normal return measure)
ARCH Model (Spot Returns)
Event OPEC Increase OPEC Decrease OPEC No Change
Day AR CAR AR CAR AR CAR
-20 0.363% 0.363% 0.164% 0.164% -0.369% -0.369%
-19 0.319% 1.007% -0.384% -0.227% -0.591% -0.960%
-18 0.272% 1.330% -0.083% -0.171% -0.179% -1.138%
-17 -0.831% 0.641% -0.103% -0.161% -0.119% -1.257%
-16 1.399% 2.055% -0.511% -0.408% 0.648% -0.609%
-15 0.147% 1.875% -0.472% -1.024% -0.004% -0.613%
-14 0.984% 2.967% 0.667% -0.259% -0.528% -1.142%
-13 0.333% 2.576% 0.129% -0.265% 0.363% -0.778%
-12 0.010% 2.533% -0.896% -0.963% 0.212% -0.566%
-11 -0.239% 3.013% 0.204% -0.655% 0.373% -0.193%
-10 0.607% 3.572% 1.106% 0.432% 0.455% 0.262%
-9 0.197% 3.553% 0.207% 0.204% 0.120% 0.382%
-8 -0.073% 2.885% 0.022% 0.318% -0.250% 0.133%
-7 0.849% 4.214% -0.423% -0.285% -0.338% -0.206%
-6 0.620% 4.387% 0.307% 0.176% -0.420% -0.626%
-5 -0.576% 3.367% -0.362% -0.249% -0.535% -1.161%
-4 -1.609% 1.857% 0.836% 0.598% 0.117% -1.043%
-3 0.672% 3.506% 0.352% 0.633% -0.698% -1.741%
-2 0.909% 3.920% 0.227% 0.925% 0.255% -1.486%
-1 -0.966% 2.907% 0.932% 2.249% -0.162% -1.647%
0 -0.754% 2.485% 0.332% 2.465% -0.148% -1.795%
1 0.050% 3.307% -0.529% 1.540% 0.755% -1.040%
2 -0.206% 3.044% 0.413% 1.873% -0.152% -1.193%
3 0.011% 3.600% 0.710% 2.651% -0.109% -1.302%
4 0.136% 3.638% 1.015% 3.548% -0.230% -1.532%
5 0.219% 3.904% 0.920% 4.439% -0.094% -1.627%
6 1.133% 4.550% 0.901% 5.660% -0.541% -2.167%
7 0.817% 4.635% 0.004% 5.813% -0.673% -2.840%
8 -0.704% 3.139% -0.082% 5.617% -0.509% -3.349%
9 -0.678% 2.634% 0.340% 6.194% -0.177% -3.526%
10 0.094% 2.513% -0.511% 5.368% 0.117% -3.409%
11 -0.259% 2.446% 1.252% 6.576% -0.723% -4.132%
12 0.694% 3.551% 1.436% 7.897% -0.108% -4.240%
13 -0.612% 2.830% 0.860% 8.190% -0.274% -4.514%
14 0.706% 3.792% -0.092% 7.801% 0.577% -3.937%
15 0.533% 3.830% 0.626% 8.446% 0.769% -3.167%
16 -0.449% 3.272% -0.355% 7.999% -0.968% -4.136%
17 0.235% 3.823% -1.586% 5.900% 0.142% -3.993%
18 0.080% 3.416% 1.504% 8.627% 0.259% -3.734%
19 -0.094% 2.944% 0.453% 9.116% -0.465% -4.199%
20 -0.995% 1.436% 0.016% 9.122% 0.627% -3.572%
Note: AR is the sample average abnormal return for the specified day in event time and CAR is the
sample average cumulative abnormal return for day -20 to the specified day. Event time is days
relative to the announcement date.
30
(Table 6 continued)
ARCH Model (Nearby Futures Contract, F1)
Event OPEC Increase OPEC Decrease OPEC No Change
Day AR CAR AR CAR AR CAR
-20 0.320% 0.320% 0.151% 0.151% -0.245% -0.245%
-19 0.416% 1.073% -0.438% -0.281% -0.629% -0.874%
-18 0.055% 1.198% -0.438% -0.563% -0.113% -0.986%
-17 -0.887% 0.490% 0.086% -0.294% 0.171% -0.816%
-16 0.989% 1.594% -0.651% -0.565% 0.632% -0.183%
-15 0.003% 1.387% -0.444% -1.158% -0.489% -0.672%
-14 0.453% 2.019% 0.988% 0.052% 0.110% -0.562%
-13 0.292% 2.007% 0.079% -0.153% 0.276% -0.286%
-12 0.240% 2.240% -0.726% -0.706% -0.113% -0.399%
-11 -0.772% 1.835% 0.125% -0.463% 0.592% 0.193%
-10 0.670% 2.808% 0.932% 0.460% 0.558% 0.751%
-9 -0.103% 2.465% 0.428% 0.615% -0.113% 0.638%
-8 0.579% 2.731% 0.063% 0.756% -0.398% 0.239%
-7 0.497% 2.862% -0.817% -0.172% -0.812% -0.573%
-6 0.630% 3.345% 0.761% 0.793% 0.068% -0.505%
-5 -0.079% 2.879% -0.406% 0.240% -0.824% -1.329%
-4 -0.619% 2.110% 0.797% 1.224% 0.383% -0.946%
-3 -0.031% 2.404% 0.101% 1.235% -0.809% -1.755%
-2 1.306% 3.484% 0.516% 1.856% 0.352% -1.403%
-1 -1.315% 2.010% 0.798% 2.424% -0.218% -1.621%
0 -0.199% 2.248% 0.202% 2.593% -0.203% -1.824%
1 0.188% 3.063% -0.351% 2.005% 0.914% -0.910%
2 -0.551% 2.330% 0.612% 2.552% -0.427% -1.337%
3 -0.308% 2.740% 0.702% 3.238% -0.243% -1.580%
4 0.287% 3.118% 0.493% 3.646% -0.064% -1.644%
5 0.274% 3.309% 0.871% 4.607% -0.488% -2.132%
6 0.791% 3.741% 0.647% 5.515% -0.315% -2.448%
7 0.211% 3.597% 0.140% 5.837% -0.400% -2.847%
8 -0.277% 2.583% -0.530% 5.125% -0.323% -3.170%
9 -1.088% 1.532% 0.341% 5.885% -0.095% -3.265%
10 0.159% 1.879% -0.349% 5.254% -0.077% -3.342%
11 0.352% 2.366% 1.124% 6.509% -0.302% -3.644%
12 0.983% 3.370% 1.494% 7.904% -0.198% -3.842%
13 -0.375% 2.705% 0.631% 7.951% -0.155% -3.997%
14 0.480% 3.029% -0.005% 7.733% 0.646% -3.351%
15 0.275% 2.975% 0.396% 8.260% 0.760% -2.591%
16 -0.173% 2.762% 0.130% 8.305% -1.064% -3.655%
17 0.573% 3.206% -0.717% 7.460% 0.006% -3.649%
18 0.261% 3.097% 0.552% 8.335% 0.358% -3.290%
19 -0.688% 2.200% 0.599% 9.130% -0.601% -3.891%
20 -0.537% 1.260% 0.204% 9.178% 0.831% -3.060%
31
(Table 6 continued)
ARCH Model (Third Closest Contract, F3)
Event OPEC Increase OPEC Decrease OPEC No Change
Day AR CAR AR CAR AR CAR
-20 0.239% 0.239% 0.204% 0.204% -0.212% -0.212%
-19 0.356% 0.838% 0.054% 0.043% -0.483% -0.696%
-18 -0.092% 0.741% -0.195% -0.442% -0.092% -0.788%
-17 -0.795% 0.255% 0.173% -0.320% 0.062% -0.726%
-16 0.815% 1.205% -0.466% -0.497% 0.516% -0.210%
-15 -0.038% 1.015% -0.119% -0.817% -0.455% -0.665%
-14 0.316% 1.583% 0.828% 0.000% -0.016% -0.681%
-13 0.366% 1.670% 0.269% -0.032% 0.198% -0.483%
-12 -0.121% 1.479% -0.455% -0.465% -0.047% -0.529%
-11 -0.857% 1.144% 0.167% -0.300% 0.526% -0.003%
-10 0.510% 2.026% 0.979% 0.635% 0.467% 0.464%
-9 -0.180% 1.735% 0.605% 0.872% -0.239% 0.225%
-8 0.497% 2.024% 0.059% 0.922% -0.080% 0.145%
-7 0.449% 2.163% -0.569% 0.221% -0.677% -0.532%
-6 0.597% 2.667% 0.477% 0.794% 0.203% -0.329%
-5 0.047% 2.390% -0.332% 0.391% -0.784% -1.113%
-4 -0.471% 1.694% 0.816% 1.329% 0.421% -0.692%
-3 -0.148% 1.823% 0.056% 1.164% -0.633% -1.325%
-2 1.302% 3.128% 0.549% 1.767% 0.176% -1.149%
-1 -1.392% 1.460% 0.757% 2.235% -0.298% -1.447%
0 0.083% 2.001% 0.188% 2.214% -0.190% -1.637%
1 0.057% 2.299% -0.054% 1.897% 0.818% -0.819%
2 -0.477% 1.895% 0.500% 2.220% -0.423% -1.242%
3 -0.240% 2.269% 0.711% 2.888% -0.193% -1.435%
4 0.253% 2.644% 0.426% 3.141% -0.078% -1.513%
5 0.165% 2.715% 0.854% 3.987% -0.394% -1.907%
6 0.580% 2.996% 0.572% 4.732% -0.236% -2.143%
7 0.001% 2.772% 0.203% 5.020% -0.428% -2.571%
8 -0.301% 1.990% -0.356% 4.395% -0.284% -2.855%
9 -0.961% 1.162% 0.315% 5.039% -0.078% -2.933%
10 0.027% 1.429% -0.284% 4.497% -0.214% -3.147%
11 0.325% 2.086% 0.994% 5.592% -0.305% -3.452%
12 0.915% 2.935% 1.457% 6.948% -0.170% -3.622%
13 -0.214% 2.425% 0.568% 6.916% -0.111% -3.733%
14 0.394% 2.638% 0.115% 6.791% 0.582% -3.150%
15 0.029% 2.365% 0.538% 7.405% 0.615% -2.535%
16 -0.209% 2.129% 0.120% 7.386% -0.738% -3.273%
17 0.547% 2.574% -0.482% 6.725% -0.016% -3.289%
18 0.076% 2.289% 0.566% 7.401% 0.225% -3.064%
19 -0.758% 1.517% 0.565% 7.997% -0.450% -3.513%
20 -0.219% 1.113% 0.219% 7.967% 0.528% -2.985%
32
(Table 6 continued)
ARCH Model (Twelfth Closest Contract, F12)
Event OPEC Increase OPEC Decrease OPEC No Change
Day AR CAR AR CAR AR CAR
-20 0.227% 0.227% -0.126% -0.126% -0.230% -0.230%
-19 0.256% 0.601% -0.105% -0.231% -0.208% -0.439%
-18 0.079% 0.657% -0.182% -0.412% 0.030% -0.408%
-17 -0.279% 0.460% 0.590% 0.178% -0.192% -0.600%
-16 0.914% 1.274% -0.242% -0.064% 0.231% -0.369%
-15 0.394% 1.476% 0.052% -0.012% -0.309% -0.678%
-14 0.014% 1.337% 0.447% 0.435% -0.100% -0.778%
-13 0.297% 1.435% 0.315% 0.750% 0.135% -0.644%
-12 -0.393% 1.044% -0.408% 0.342% -0.090% -0.734%
-11 -0.605% 0.872% 0.023% 0.365% 0.485% -0.249%
-10 0.532% 1.584% 0.426% 0.791% 0.259% 0.010%
-9 -0.311% 1.309% 0.694% 1.485% -0.371% -0.361%
-8 0.297% 1.566% 0.014% 1.499% -0.140% -0.501%
-7 0.600% 2.160% -0.575% 0.924% -0.426% -0.927%
-6 0.558% 2.514% 0.339% 1.263% 0.425% -0.501%
-5 0.073% 2.494% -0.295% 0.968% -0.602% -1.103%
-4 -0.249% 2.041% 0.478% 1.446% 0.564% -0.539%
-3 0.100% 2.280% 0.066% 1.512% -0.402% -0.941%
-2 1.320% 3.402% 0.273% 1.785% 0.226% -0.715%
-1 -1.254% 1.838% 0.640% 2.425% -0.404% -1.119%
0 -0.068% 1.989% -0.251% 2.174% -0.200% -1.319%
1 0.329% 2.841% -0.255% 1.919% 0.540% -0.779%
2 -0.224% 2.644% 0.399% 2.318% -0.525% -1.305%
3 -0.074% 2.972% 0.497% 2.816% -0.047% -1.351%
4 0.254% 3.158% 0.067% 2.883% 0.115% -1.236%
5 -0.171% 3.079% 1.131% 4.013% -0.335% -1.571%
6 0.314% 3.396% 0.404% 4.417% -0.212% -1.783%
7 -0.330% 3.124% -0.305% 4.112% -0.386% -2.169%
8 -0.318% 2.603% 0.018% 4.130% -0.171% -2.340%
9 -0.227% 2.628% 0.007% 4.137% 0.044% -2.296%
10 0.128% 2.512% -0.212% 3.925% -0.308% -2.603%
11 0.245% 3.081% 0.784% 4.708% -0.191% -2.794%
12 0.551% 3.528% 0.764% 5.473% -0.080% -2.874%
13 -0.164% 3.201% -0.014% 5.459% 0.033% -2.841%
14 0.233% 3.360% 0.303% 5.763% 0.601% -2.239%
15 -0.239% 3.007% 0.507% 6.270% 0.602% -1.637%
16 -0.455% 2.494% -0.184% 6.086% -0.602% -2.239%
17 0.164% 2.752% -0.628% 5.458% 0.089% -2.150%
18 0.217% 3.082% 0.571% 6.029% 0.149% -2.001%
19 -0.784% 2.263% 0.177% 6.206% -0.213% -2.214%
20 0.037% 2.118% 0.313% 6.519% 0.413% -1.801%
33
Table 7. Abnormal and Cumulative Abnormal Returns Around OPEC Announcements. (The abnormal
return is calculated using the Fama-French model as the normal return measure)
Fama-French Model (Spot Returns)
Event OPEC Increase OPEC Decrease OPEC No Change
Day AR CAR AR CAR AR CAR
-20 0.858% 0.858% 0.795% 0.795% -0.311% -0.311%
-19 0.107% 0.938% -0.533% -0.120% -0.443% -0.753%
-18 0.563% 1.641% 0.200% 0.235% -0.081% -0.834%
-17 -1.069% 0.559% -0.976% -0.656% 0.416% -0.419%
-16 1.172% 1.800% -0.301% -0.679% 0.968% 0.549%
-15 -0.327% 1.189% -0.557% -1.408% -0.327% 0.222%
-14 0.397% 2.004% 0.245% -1.020% -0.551% -0.329%
-13 0.115% 1.814% -0.148% -1.209% 0.192% -0.136%
-12 -0.452% 1.159% -1.406% -2.601% -0.049% -0.186%
-11 -0.103% 1.942% -0.304% -2.690% 0.333% 0.148%
-10 0.515% 2.388% 0.375% -2.374% 0.044% 0.192%
-9 -0.213% 2.048% 0.783% -1.967% 0.077% 0.269%
-8 -0.160% 1.443% 0.278% -1.671% -0.132% 0.137%
-7 1.323% 3.063% -0.075% -2.057% -0.372% -0.235%
-6 0.766% 3.166% 0.312% -1.485% -0.257% -0.492%
-5 -0.433% 2.123% -0.187% -1.743% -0.796% -1.288%
-4 -1.269% 0.880% 0.899% -0.859% -0.119% -1.407%
-3 0.633% 2.367% 0.091% -1.163% -1.035% -2.442%
-2 1.099% 2.565% -0.058% -1.317% 0.451% -1.991%
-1 -1.251% 0.889% 1.177% 0.298% -0.583% -2.574%
0 -0.727% 0.862% 0.382% 0.528% -0.441% -3.015%
1 -0.065% 1.623% -0.685% -0.506% 0.904% -2.111%
2 -0.578% 1.154% -0.073% -0.626% 0.054% -2.057%
3 0.276% 2.161% 0.258% -0.209% -0.239% -2.296%
4 0.212% 2.030% 1.012% 0.842% -0.236% -2.532%
5 -0.084% 1.891% 0.367% 1.056% -0.109% -2.641%
6 1.263% 2.940% 0.384% 1.704% -0.538% -3.179%
7 0.781% 2.793% -0.650% 1.692% -0.703% -3.882%
8 -0.806% 1.071% -0.229% 1.576% -0.357% -4.239%
9 -0.787% 0.524% 0.386% 2.353% -0.314% -4.552%
10 -0.462% -0.207% -0.301% 1.839% -0.034% -4.586%
11 -0.476% -0.201% 0.508% 2.528% -0.602% -5.188%
12 0.432% 0.608% 0.861% 3.613% -0.686% -5.874%
13 -0.807% -0.258% 0.514% 3.970% -0.466% -6.340%
14 0.352% 0.327% 0.277% 4.124% 0.967% -5.373%
15 0.476% 0.488% 0.253% 4.425% 0.173% -5.199%
16 -0.068% 0.297% -0.742% 3.599% -1.032% -6.231%
17 -0.079% 0.299% -2.090% 1.148% 0.218% -6.013%
18 0.065% -0.088% 1.687% 3.897% -0.031% -6.044%
19 0.251% -0.178% -0.152% 3.748% -1.058% -7.102%
20 -0.694% -1.593% -0.141% 3.642% 0.719% -6.383%
34
(Table 7 continued)
Fama-French Model (Nearby Futures Contract, F1)
Event OPEC Increase OPEC Decrease OPEC No Change
Day AR CAR AR CAR AR CAR
-20 0.520% 0.520% 0.886% 0.886% -0.162% -0.162%
-19 0.152% 0.750% -0.775% -0.092% -0.523% -0.686%
-18 0.218% 1.131% -0.332% -0.292% -0.002% -0.687%
-17 -1.228% 0.064% -0.537% -0.632% 0.637% -0.050%
-16 0.878% 1.117% -0.381% -0.622% 0.865% 0.815%
-15 -0.417% 0.502% -0.569% -1.371% -0.722% 0.092%
-14 0.380% 1.369% 0.497% -0.690% 0.040% 0.132%
-13 0.073% 1.114% -0.390% -1.223% 0.178% 0.310%
-12 -0.199% 0.770% -1.262% -2.254% -0.351% -0.041%
-11 -0.816% 0.523% -0.389% -2.412% 0.680% 0.639%
-10 0.784% 1.748% 0.454% -1.929% 0.178% 0.817%
-9 -0.367% 1.179% 0.811% -1.451% -0.117% 0.700%
-8 0.342% 1.248% 0.333% -0.964% -0.388% 0.312%
-7 0.911% 1.570% -0.588% -1.748% -0.809% -0.496%
-6 0.919% 2.066% 0.497% -0.954% 0.062% -0.434%
-5 0.074% 1.555% -0.292% -1.247% -1.073% -1.507%
-4 -0.235% 1.072% 0.883% -0.128% 0.234% -1.273%
-3 -0.090% 1.062% -0.259% -0.545% -1.122% -2.395%
-2 1.619% 2.049% 0.083% -0.491% 0.456% -1.939%
-1 -1.552% -0.076% 0.868% 0.193% -0.541% -2.480%
0 0.050% 0.668% 0.275% 0.361% -0.510% -2.990%
1 0.117% 1.309% -0.510% -0.360% 1.106% -1.884%
2 -0.623% 0.555% 0.338% -0.068% -0.293% -2.177%
3 -0.097% 1.153% 0.304% 0.137% -0.395% -2.573%
4 0.354% 1.342% 0.288% 0.433% -0.124% -2.697%
5 0.369% 1.632% 0.474% 0.853% -0.501% -3.198%
6 0.823% 2.052% 0.153% 1.154% -0.354% -3.552%
7 0.242% 1.813% -0.435% 1.364% -0.399% -3.950%
8 -0.554% 0.282% -0.645% 0.799% -0.187% -4.137%
9 -1.096% -0.729% 0.432% 1.817% -0.175% -4.312%
10 -0.177% -0.779% -0.380% 1.296% -0.205% -4.517%
11 0.316% -0.175% 0.318% 1.978% -0.124% -4.642%
12 0.717% 0.465% 0.881% 3.216% -0.756% -5.398%
13 -0.278% -0.010% 0.243% 3.264% -0.318% -5.715%
14 -0.065% -0.143% 0.154% 3.388% 1.000% -4.715%
15 0.067% -0.072% -0.091% 3.363% 0.309% -4.406%
16 0.114% -0.080% -0.325% 2.921% -1.217% -5.624%
17 0.252% -0.152% -1.361% 1.580% 0.130% -5.494%
18 0.369% -0.183% 0.534% 2.296% 0.006% -5.488%
19 -0.466% -0.911% 0.421% 2.853% -1.028% -6.516%
20 -0.298% -1.855% 0.195% 2.852% 0.851% -5.665%
35
(Table 7 continued)
Fama-French Model (Third Closest Contract, F3)
Event OPEC Increase OPEC Decrease OPEC No Change
Day AR CAR AR CAR AR CAR
-20 0.336% 0.336% 0.626% 0.626% -0.153% -0.153%
-19 0.137% 0.553% -0.193% 0.114% -0.400% -0.553%
-18 -0.037% 0.599% -0.026% -0.207% 0.011% -0.542%
-17 -1.128% -0.151% -0.302% -0.613% 0.451% -0.091%
-16 0.724% 0.801% -0.269% -0.564% 0.713% 0.622%
-15 -0.180% 0.503% -0.247% -0.993% -0.681% -0.059%
-14 0.583% 1.383% 0.441% -0.539% -0.100% -0.159%
-13 0.145% 0.923% -0.150% -0.855% 0.082% -0.078%
-12 -0.417% 0.380% -0.881% -1.645% -0.262% -0.340%
-11 -0.910% 0.124% -0.327% -1.828% 0.594% 0.255%
-10 0.799% 1.438% 0.637% -1.117% 0.125% 0.380%
-9 -0.217% 0.978% 0.921% -0.590% -0.226% 0.154%
-8 0.289% 0.905% 0.250% -0.324% 0.002% 0.155%
-7 0.765% 1.165% -0.220% -0.755% -0.658% -0.503%
-6 0.872% 1.746% 0.291% -0.444% 0.211% -0.292%
-5 0.243% 1.494% -0.245% -0.673% -0.944% -1.236%
-4 -0.100% 1.044% 0.908% 0.384% 0.347% -0.889%
-3 -0.224% 0.859% -0.327% -0.205% -0.963% -1.852%
-2 1.636% 2.179% 0.232% 0.056% 0.215% -1.637%
-1 -1.536% -0.028% 0.852% 0.625% -0.513% -2.150%
0 0.473% 1.060% 0.255% 0.572% -0.470% -2.620%
1 0.058% 1.064% -0.170% 0.131% 0.918% -1.702%
2 -0.467% 0.640% 0.383% 0.336% -0.343% -2.045%
3 -0.049% 1.112% 0.425% 0.562% -0.360% -2.405%
4 0.323% 1.342% 0.137% 0.552% -0.114% -2.519%
5 0.467% 1.718% 0.520% 1.126% -0.412% -2.932%
6 0.563% 1.749% 0.383% 1.656% -0.247% -3.179%
7 0.093% 1.569% -0.233% 1.717% -0.407% -3.585%
8 -0.542% 0.311% -0.511% 1.135% -0.151% -3.737%
9 -0.998% -0.498% 0.452% 2.133% -0.200% -3.937%
10 -0.347% -0.626% -0.189% 1.717% -0.345% -4.281%
11 0.203% 0.124% 0.609% 2.479% -0.148% -4.429%
12 0.600% 0.638% 1.089% 3.592% -0.682% -5.110%
13 -0.188% 0.322% 0.273% 3.466% -0.266% -5.376%
14 -0.097% 0.201% 0.082% 3.527% 0.869% -4.507%
15 -0.077% 0.163% 0.046% 3.730% 0.291% -4.217%
16 0.031% 0.046% -0.166% 3.491% -0.897% -5.114%
17 0.249% 0.044% -0.910% 2.467% 0.097% -5.017%
18 0.182% -0.135% 0.566% 2.972% -0.073% -5.090%
19 -0.572% -0.792% 0.378% 3.344% -0.776% -5.866%
20 0.017% -1.164% 0.234% 3.411% 0.520% -5.345%
36
(Table 7 continued)
Fama-French Model (Twelfth Closest Contract, F12)
Event OPEC Increase OPEC Decrease OPEC No Change
Day AR CAR AR CAR AR CAR
-20 0.260% 0.260% 0.069% 0.069% -0.216% -0.216%
-19 0.182% 0.494% -0.254% -0.185% -0.136% -0.352%
-18 0.210% 0.661% -0.041% -0.226% 0.136% -0.216%
-17 -0.449% 0.243% 0.303% 0.077% 0.080% -0.136%
-16 0.948% 1.077% -0.096% -0.019% 0.386% 0.250%
-15 0.294% 1.130% 0.061% 0.041% -0.487% -0.237%
-14 0.021% 0.998% 0.355% 0.397% -0.211% -0.448%
-13 0.125% 0.869% 0.025% 0.422% 0.053% -0.395%
-12 -0.464% 0.437% -0.588% -0.166% -0.241% -0.636%
-11 -0.577% 0.282% -0.304% -0.470% 0.438% -0.197%
-10 0.729% 1.202% 0.168% -0.302% -0.057% -0.255%
-9 -0.445% 0.724% 0.973% 0.671% -0.376% -0.630%
-8 0.024% 0.665% 0.090% 0.761% -0.051% -0.681%
-7 0.718% 1.371% -0.364% 0.397% -0.383% -1.064%
-6 0.681% 1.787% 0.334% 0.730% 0.399% -0.665%
-5 0.164% 1.809% -0.152% 0.579% -0.756% -1.422%
-4 -0.070% 1.497% 0.514% 1.093% 0.546% -0.875%
-3 -0.107% 1.502% -0.245% 0.848% -0.726% -1.602%
-2 1.311% 2.524% 0.030% 0.878% 0.244% -1.358%
-1 -1.418% 0.776% 0.660% 1.538% -0.460% -1.818%
0 -0.122% 0.960% -0.246% 1.292% -0.418% -2.236%
1 0.322% 1.819% -0.375% 0.917% 0.567% -1.669%
2 -0.242% 1.536% 0.228% 1.145% -0.494% -2.162%
3 0.028% 1.901% 0.303% 1.448% -0.179% -2.342%
4 0.262% 2.015% -0.114% 1.333% 0.088% -2.253%
5 -0.095% 1.997% 0.904% 2.238% -0.428% -2.681%
6 0.309% 2.306% 0.476% 2.714% -0.227% -2.909%
7 -0.278% 2.033% -0.448% 2.266% -0.298% -3.207%
8 -0.435% 1.303% 0.086% 2.352% -0.015% -3.221%
9 -0.246% 1.280% 0.089% 2.441% -0.120% -3.342%
10 -0.041% 0.929% -0.106% 2.335% -0.436% -3.778%
11 0.018% 1.237% 0.657% 2.992% -0.120% -3.898%
12 0.373% 1.482% 0.690% 3.682% -0.497% -4.395%
13 -0.042% 1.232% 0.004% 3.686% -0.150% -4.545%
14 0.148% 1.306% 0.348% 4.035% 0.733% -3.812%
15 -0.320% 0.863% 0.166% 4.201% 0.457% -3.354%
16 -0.240% 0.567% -0.386% 3.814% -0.676% -4.031%
17 0.023% 0.587% -1.075% 2.740% 0.198% -3.833%
18 0.209% 0.848% 0.540% 3.280% -0.085% -3.918%
19 -0.761% 0.030% 0.171% 3.450% -0.394% -4.312%
20 0.146% -0.019% 0.263% 3.713% 0.448% -3.864%
Table 8. Abnormal and Cumulative Abnormal Returns Around SPR Announcements. (The abnormal
return is calculated using the market model as the normal return measure)
Market Model (Spot Returns) Market Model (Nearby Futures Contract, F1)
37
Event SPR Increase SPR Decrease SPR Increase SPR Decrease
Day AR CAR AR CAR AR CAR AR CAR
-20 -0.757% -0.757% -0.037% -0.037% 0.526% 0.526% -0.127% -0.127%
-19 1.178% 0.421% -0.604% -0.641% 0.230% 0.756% 0.090% -0.037%
-18 -0.513% -0.091% -1.626% -2.267% -1.529% -0.773% -1.458% -1.495%
-17 -0.961% -1.053% -0.739% -3.006% -0.064% -0.837% -0.771% -2.266%
-16 -0.487% -1.540% 0.178% -2.828% -0.228% -1.065% 0.373% -1.894%
-15 -0.059% -1.599% 1.167% -1.661% -0.671% -1.736% 1.011% -0.883%
-14 -0.164% -1.762% -0.938% -2.599% 0.643% -1.093% -0.505% -1.388%
-13 -0.054% -1.816% 0.244% -2.355% -0.923% -2.016% 0.214% -1.173%
-12 -0.672% -2.488% -0.117% -2.471% -0.248% -2.264% -0.411% -1.585%
-11 0.061% -2.428% 0.883% -1.588% 0.939% -1.325% 0.407% -1.178%
-10 1.033% -1.394% -0.092% -1.681% 0.838% -0.487% -0.852% -2.030%
-9 -1.917% -3.312% -0.059% -1.740% -1.018% -1.505% -0.675% -2.705%
-8 0.285% -3.026% -1.680% -3.419% -0.687% -2.192% -1.135% -3.840%
-7 -0.975% -4.001% 0.130% -3.289% -0.651% -2.843% -0.047% -3.887%
-6 -1.004% -5.005% -0.020% -3.309% -0.923% -3.766% 0.638% -3.249%
-5 -0.862% -5.867% -0.312% -3.621% -0.696% -4.462% 0.509% -2.739%
-4 0.263% -5.604% -0.430% -4.051% 0.554% -3.908% -0.398% -3.137%
-3 -0.513% -6.116% 0.137% -3.914% 0.376% -3.532% -0.291% -3.428%
-2 -1.118% -7.234% -1.223% -5.137% -0.596% -4.127% 0.562% -2.866%
-1 -0.488% -7.723% 0.514% -4.624% -0.539% -4.666% 0.703% -2.163%
0 -1.228% -8.951% 0.915% -3.709% -1.723% -6.389% 0.134% -2.029%
1 0.943% -8.008% -0.621% -4.330% 0.464% -5.925% -0.544% -2.573%
2 -0.833% -8.841% 0.336% -3.994% -0.331% -6.255% 0.390% -2.182%
3 -0.997% -9.838% -1.336% -5.330% -0.359% -6.614% -1.506% -3.688%
4 1.788% -8.049% -0.424% -5.753% 1.115% -5.499% -0.390% -4.078%
5 0.340% -7.709% -0.774% -6.527% 0.565% -4.934% -0.690% -4.768%
6 0.553% -7.156% 0.171% -6.356% 0.375% -4.560% -0.006% -4.774%
7 0.067% -7.089% -1.108% -7.464% 0.250% -4.309% -0.675% -5.449%
8 0.406% -6.683% 0.873% -6.591% 0.537% -3.773% 0.865% -4.585%
9 -0.744% -7.427% -0.265% -6.856% -0.649% -4.422% -0.289% -4.874%
10 -0.639% -8.066% 0.229% -6.628% -0.442% -4.864% 0.200% -4.673%
11 0.086% -7.981% 0.560% -6.067% 0.102% -4.762% 0.379% -4.295%
12 0.514% -7.467% -0.219% -6.286% 0.473% -4.289% -0.208% -4.503%
13 -0.088% -7.554% 0.782% -5.504% 0.189% -4.100% 0.109% -4.394%
14 -1.377% -8.931% 0.749% -4.754% -0.753% -4.853% 0.357% -4.037%
15 0.830% -8.101% 0.849% -3.906% 0.403% -4.450% -0.049% -4.086%
16 -0.319% -8.420% -0.409% -4.314% -0.126% -4.576% 0.188% -3.898%
17 0.689% -7.731% -1.488% -5.802% -0.154% -4.730% -1.198% -5.096%
18 0.033% -7.698% -1.185% -6.988% 0.582% -4.147% -0.210% -5.307%
19 -1.269% -8.967% -0.248% -7.236% -1.340% -5.488% -0.022% -5.329%
20 -0.625% -9.592% -1.067% -8.302% 0.177% -5.310% -1.170% -6.499%
38
(Table 8 continued)
Market Model (Third Closest Contract, F3) Market Model (Twelfth Closest Contract, F12)
Event SPR Increase SPR Decrease SPR Increase SPR Decrease
Day AR CAR AR CAR AR CAR AR CAR
-20 0.427% 0.427% 0.161% 0.161% 0.379% 0.379% 0.419% 0.419%
-19 0.423% 0.850% 0.126% 0.288% 0.376% 0.755% -0.072% 0.347%
-18 -1.097% -0.248% -1.162% -0.874% -0.715% 0.040% -1.025% -0.677%
-17 -0.059% -0.307% -0.564% -1.438% 0.058% 0.098% -0.080% -0.758%
-16 -0.218% -0.525% 0.184% -1.254% -0.472% -0.373% -0.173% -0.930%
-15 -0.351% -0.876% 0.801% -0.453% -0.321% -0.695% 0.570% -0.361%
-14 0.617% -0.259% -0.401% -0.854% 0.604% -0.090% -0.253% -0.613%
-13 -0.852% -1.111% 0.154% -0.700% -0.806% -0.896% 0.019% -0.594%
-12 -0.251% -1.361% -0.379% -1.079% -0.238% -1.135% -0.372% -0.966%
-11 0.829% -0.533% -0.077% -1.156% 0.805% -0.330% -0.636% -1.602%
-10 0.636% 0.103% -0.947% -2.103% 0.122% -0.209% -0.860% -2.463%
-9 -0.949% -0.845% -0.525% -2.627% -0.696% -0.905% -0.447% -2.910%
-8 -0.662% -1.508% -0.538% -3.165% -0.464% -1.369% -0.396% -3.306%
-7 -0.963% -2.471% -0.156% -3.321% -1.101% -2.469% -0.126% -3.432%
-6 -0.880% -3.350% 0.726% -2.595% -0.923% -3.393% 0.570% -2.862%
-5 -0.613% -3.964% 0.224% -2.371% -0.578% -3.971% -0.506% -3.368%
-4 0.601% -3.363% -0.203% -2.573% 0.447% -3.524% -0.005% -3.373%
-3 0.588% -2.775% -0.016% -2.589% 0.535% -2.990% 0.164% -3.208%
-2 -0.583% -3.358% 0.677% -1.912% -0.202% -3.192% 0.515% -2.694%
-1 -0.223% -3.581% 0.608% -1.304% 0.007% -3.185% 0.103% -2.590%
0 -1.705% -5.286% 0.264% -1.040% -1.352% -4.538% -0.141% -2.731%
1 0.095% -5.191% -0.482% -1.521% 0.596% -3.942% -0.489% -3.220%
2 -0.129% -5.321% 0.477% -1.045% -0.156% -4.098% 0.195% -3.025%
3 -0.436% -5.757% -1.191% -2.235% -0.326% -4.424% -0.456% -3.480%
4 0.705% -5.052% -0.260% -2.495% 0.254% -4.170% -0.331% -3.812%
5 0.542% -4.510% -0.449% -2.944% 0.337% -3.833% -0.092% -3.904%
6 0.262% -4.248% 0.138% -2.806% 0.281% -3.552% 0.161% -3.743%
7 0.355% -3.893% -0.604% -3.410% 0.341% -3.211% -0.416% -4.159%
8 0.631% -3.262% 0.796% -2.614% 0.806% -2.405% 0.602% -3.557%
9 -0.587% -3.850% -0.186% -2.800% -0.449% -2.854% -0.054% -3.611%
10 -0.501% -4.351% 0.161% -2.639% -0.462% -3.316% 0.074% -3.537%
11 0.244% -4.107% 0.284% -2.355% 0.254% -3.061% 0.209% -3.328%
12 0.349% -3.758% -0.123% -2.478% 0.201% -2.861% 0.049% -3.279%
13 0.144% -3.614% -0.001% -2.479% 0.020% -2.840% 0.209% -3.070%
14 -0.653% -4.267% 0.157% -2.322% -0.487% -3.328% 0.112% -2.958%
15 0.580% -3.687% -0.355% -2.677% 0.572% -2.755% -0.710% -3.667%
16 -0.248% -3.935% 0.154% -2.523% -0.277% -3.032% 0.107% -3.560%
17 -0.085% -4.020% -1.067% -3.589% -0.110% -3.142% -0.849% -4.409%
18 0.450% -3.570% -0.093% -3.682% 0.386% -2.756% 0.101% -4.308%
19 -0.855% -4.425% -0.107% -3.789% -0.535% -3.290% -0.238% -4.547%
20 0.418% -4.007% -1.085% -4.874% 0.583% -2.708% -0.745% -5.291%
Table 9. Abnormal and Cumulative Abnormal Returns Around SPR Announcements. (The abnormal
return is calculated using the ARCH model as the normal return measure)
39
ARCH Model (Spot Returns) ARCH Model (Nearby Futures Contract, F1)
Event SPR Increase SPR Decrease SPR Increase SPR Decrease
Day AR CAR AR CAR AR CAR AR CAR
-20 -0.203% -0.203% 0.982% 0.982% 1.088% 1.088% 0.872% 0.872%
-19 0.105% -0.097% 0.286% 1.268% -0.688% 0.400% 1.028% 1.900%
-18 0.666% 0.569% -1.285% -0.017% -0.402% -0.002% -1.308% 0.591%
-17 -0.257% 0.312% 0.034% 0.018% 0.820% 0.817% 0.011% 0.603%
-16 -1.029% -0.717% 0.041% 0.058% -0.827% -0.009% 0.794% 1.397%
-15 0.020% -0.697% 1.909% 1.967% -0.770% -0.780% 1.930% 3.326%
-14 -0.122% -0.819% -0.720% 1.247% 0.621% -0.158% -0.367% 2.960%
-13 -0.157% -0.976% 1.367% 2.614% -0.875% -1.033% 1.214% 4.174%
-12 -0.641% -1.617% -0.023% 2.591% -0.232% -1.265% 0.015% 4.189%
-11 0.082% -1.535% 0.063% 2.654% 0.982% -0.284% -0.207% 3.982%
-10 0.176% -1.360% -2.155% 0.499% 0.054% -0.230% -2.320% 1.662%
-9 -2.778% -4.138% 0.036% 0.535% -1.792% -2.022% -0.343% 1.318%
-8 0.805% -3.333% 0.406% 0.941% -0.041% -2.062% 0.833% 2.151%
-7 -1.510% -4.843% -0.989% -0.048% -1.189% -3.251% -1.104% 1.047%
-6 -1.249% -6.093% 1.510% 1.462% -1.249% -4.501% 1.787% 2.834%
-5 -0.950% -7.043% 0.755% 2.217% -0.880% -5.380% 1.338% 4.173%
-4 1.680% -5.363% -0.338% 1.878% 1.769% -3.611% -0.182% 3.991%
-3 -0.436% -5.799% 2.851% 4.730% -0.004% -3.615% 2.801% 6.792%
-2 -0.036% -5.835% -0.869% 3.860% 0.607% -3.007% 0.510% 7.302%
-1 -0.112% -5.947% 1.044% 4.905% -0.498% -3.506% 1.034% 8.336%
0 -0.283% -6.229% -6.492% -1.587% -0.920% -4.425% -6.449% 1.887%
1 -2.267% -8.496% -1.372% -2.959% -2.153% -6.578% -0.958% 0.929%
2 0.848% -7.648% 1.617% -1.342% 1.279% -5.300% 1.207% 2.136%
3 -0.569% -8.218% 2.537% 1.195% -0.001% -5.301% 1.117% 3.253%
4 1.189% -7.029% -1.226% -0.031% 0.446% -4.855% -0.362% 2.891%
5 1.495% -5.534% 0.611% 0.580% 1.613% -3.242% -0.418% 2.474%
6 -0.307% -5.841% -0.736% -0.156% -0.578% -3.820% -0.691% 1.782%
7 0.682% -5.158% -3.010% -3.166% 0.738% -3.081% -0.134% 1.648%
8 -0.201% -5.359% 1.166% -2.000% -0.020% -3.101% 1.253% 2.901%
9 -0.126% -5.485% -0.697% -2.696% 0.001% -3.100% -0.815% 2.086%
10 1.302% -4.184% 1.056% -1.640% 1.250% -1.850% 0.924% 3.010%
11 1.136% -3.048% 0.410% -1.230% 1.279% -0.571% 0.607% 3.617%
12 0.439% -2.609% -1.597% -2.827% 0.533% -0.038% -1.413% 2.203%
13 -0.106% -2.715% -1.155% -3.982% 0.184% 0.146% -1.306% 0.898%
14 -2.144% -4.859% 3.364% -0.618% -1.172% -1.026% 2.799% 3.697%
15 1.094% -3.765% 0.155% -0.463% 0.475% -0.552% -0.486% 3.211%
16 -1.660% -5.425% 0.558% 0.095% -1.346% -1.898% 1.017% 4.228%
17 0.782% -4.643% -0.275% -0.180% -0.196% -2.094% 0.112% 4.340%
18 0.169% -4.474% -0.591% -0.771% 0.730% -1.364% 0.102% 4.441%
19 -1.708% -6.182% -0.884% -1.655% -1.741% -3.105% -0.421% 4.020%
20 1.532% -4.649% -1.810% -3.465% 1.987% -1.118% -1.745% 2.275%
40
(Table 9 continued)
ARCH Model (Third Closest Contract, F3) ARCH Model (Twelfth Closest Contract, F12)
Event SPR Increase SPR Decrease SPR Increase SPR Decrease
Day AR CAR AR CAR AR CAR AR CAR
-20 0.938% 0.938% 0.863% 0.863% 0.726% 0.726% 0.963% 0.963%
-19 -0.309% 0.629% 0.813% 1.676% -0.172% 0.555% 0.530% 1.492%
-18 -0.128% 0.501% -0.814% 0.861% 0.002% 0.557% -0.787% 0.706%
-17 0.766% 1.266% -0.030% 0.831% 0.650% 1.207% 0.129% 0.835%
-16 -0.720% 0.546% 0.530% 1.361% -0.850% 0.356% 0.179% 1.014%
-15 -0.466% 0.080% 1.555% 2.916% -0.406% -0.049% 1.207% 2.221%
-14 0.655% 0.734% -0.431% 2.485% 0.602% 0.553% -0.365% 1.856%
-13 -0.838% -0.103% 1.014% 3.499% -0.854% -0.301% 0.580% 2.436%
-12 -0.261% -0.364% 0.102% 3.601% -0.269% -0.570% 0.082% 2.518%
-11 0.873% 0.509% -0.476% 3.126% 0.816% 0.246% -0.996% 1.523%
-10 -0.011% 0.498% -1.984% 1.142% -0.366% -0.120% -1.353% 0.170%
-9 -1.583% -1.085% -0.222% 0.920% -1.181% -1.301% -0.077% 0.093%
-8 -0.053% -1.138% 0.604% 1.524% 0.008% -1.293% 0.823% 0.916%
-7 -1.404% -2.542% -0.731% 0.793% -1.440% -2.732% -0.846% 0.070%
-6 -1.136% -3.678% 1.545% 2.338% -1.077% -3.810% 1.085% 1.155%
-5 -0.815% -4.493% 0.733% 3.070% -0.740% -4.549% -0.470% 0.686%
-4 1.633% -2.860% -0.033% 3.037% 1.186% -3.363% 0.051% 0.737%
-3 0.168% -2.692% 1.847% 4.884% 0.180% -3.183% 1.502% 2.239%
-2 0.536% -2.156% 1.014% 5.898% 0.658% -2.526% 0.532% 2.771%
-1 -0.245% -2.401% 0.587% 6.485% -0.014% -2.540% 0.064% 2.835%
0 -0.953% -3.354% -4.522% 1.963% -0.784% -3.323% -2.147% 0.689%
1 -1.911% -5.264% -0.765% 1.198% -0.911% -4.234% -0.333% 0.356%
2 1.190% -4.074% 0.924% 2.122% 0.805% -3.430% 0.594% 0.950%
3 -0.130% -4.204% 0.774% 2.897% -0.086% -3.516% 0.766% 1.716%
4 0.113% -4.091% -0.451% 2.445% -0.234% -3.750% -0.593% 1.123%
5 1.482% -2.609% -0.492% 1.953% 1.071% -2.679% -0.289% 0.834%
6 -0.498% -3.107% -0.496% 1.457% -0.289% -2.968% -0.437% 0.397%
7 0.704% -2.403% 0.056% 1.513% 0.592% -2.376% 0.413% 0.810%
8 0.219% -2.183% 1.099% 2.612% 0.489% -1.887% 0.972% 1.782%
9 0.013% -2.170% -0.733% 1.880% -0.016% -1.902% -0.618% 1.163%
10 0.876% -1.295% 0.795% 2.674% 0.551% -1.351% 0.712% 1.875%
11 1.238% -0.056% 0.499% 3.173% 0.998% -0.354% 0.525% 2.401%
12 0.487% 0.430% -1.094% 2.079% 0.320% -0.034% -0.591% 1.810%
13 0.180% 0.610% -1.167% 0.912% 0.053% 0.018% -0.691% 1.119%
14 -0.978% -0.368% 2.154% 3.066% -0.747% -0.729% 1.590% 2.709%
15 0.644% 0.277% -0.658% 2.408% 0.644% -0.085% -0.722% 1.987%
16 -1.253% -0.976% 0.606% 3.014% -1.028% -1.113% 0.360% 2.347%
17 -0.097% -1.074% 0.174% 3.189% -0.129% -1.242% 0.140% 2.487%
18 0.600% -0.474% 0.140% 3.328% 0.519% -0.723% 0.195% 2.682%
19 -1.218% -1.691% -0.515% 2.813% -0.801% -1.524% -0.488% 2.194%
20 1.930% 0.238% -1.415% 1.398% 1.709% 0.185% -0.807% 1.387%
41
Table 10. Abnormal and Cumulative Abnormal Returns Around SPR Announcements. (The abnormal
return is calculated using the Fama-French model as the normal return measure)
Fama-French Model (Spot Returns) Fama-French Model (Nearby Futures Contract, F1)
Event SPR Increase SPR Decrease SPR Increase SPR Decrease
Day AR CAR AR CAR AR CAR AR CAR
-20 -0.375% -0.375% 0.643% 0.643% 0.942% 0.942% 0.358% 0.358%
-19 0.058% -0.317% 0.615% 1.258% -0.803% 0.139% 1.152% 1.511%
-18 1.283% 0.967% -1.140% 0.118% 0.644% 0.783% -1.133% 0.378%
-17 0.748% 1.715% -0.094% 0.024% 1.519% 2.302% -0.202% 0.176%
-16 -0.686% 1.029% 0.149% 0.173% -0.513% 1.789% 0.886% 1.062%
-15 0.466% 1.495% 2.193% 2.367% -0.274% 1.515% 2.005% 3.067%
-14 0.295% 1.790% -1.208% 1.159% 1.134% 2.648% -0.640% 2.427%
-13 -0.008% 1.783% 0.715% 1.874% -0.524% 2.125% 0.588% 3.015%
-12 -0.919% 0.863% 0.384% 2.258% -0.395% 1.730% 0.490% 3.506%
-11 -0.616% 0.248% -0.412% 1.846% 0.761% 2.491% -0.286% 3.220%
-10 -0.092% 0.155% -2.547% -0.701% -0.070% 2.421% -2.509% 0.711%
-9 -1.783% -1.627% 0.155% -0.546% -1.140% 1.281% -0.251% 0.460%
-8 1.246% -0.381% -0.200% -0.746% -0.021% 1.260% 0.331% 0.792%
-7 -0.764% -1.146% -1.480% -2.225% -0.601% 0.660% -1.581% -0.790%
-6 -0.213% -1.359% 0.508% -1.718% -0.535% 0.125% 1.104% 0.314%
-5 -0.558% -1.917% 1.120% -0.598% -0.802% -0.677% 1.390% 1.704%
-4 1.418% -0.499% -0.474% -1.072% 1.317% 0.640% -0.267% 1.437%
-3 -0.900% -1.398% 2.006% 0.934% -1.009% -0.369% 1.790% 3.226%
-2 -1.273% -2.672% -1.104% -0.170% 0.171% -0.198% 0.109% 3.335%
-1 0.384% -2.288% 1.332% 1.162% -0.294% -0.492% 1.273% 4.608%
0 -0.454% -2.742% -4.280% -3.118% -1.049% -1.540% -4.787% -0.179%
1 -2.917% -5.659% -1.210% -4.329% -2.775% -4.315% -1.179% -1.358%
2 1.559% -4.100% 1.483% -2.846% 1.824% -2.491% 1.198% -0.160%
3 -0.354% -4.454% 2.202% -0.644% 0.587% -1.904% 1.102% 0.942%
4 0.853% -3.601% -0.601% -1.245% 0.192% -1.712% 0.402% 1.344%
5 1.481% -2.120% 1.789% 0.544% 1.346% -0.366% 0.790% 2.135%
6 -0.024% -2.144% -0.184% 0.360% -0.553% -0.920% -0.134% 2.001%
7 -0.052% -2.197% -2.671% -2.310% 0.649% -0.270% 0.379% 2.380%
8 -0.948% -3.145% 0.911% -1.400% -0.641% -0.911% 1.195% 3.575%
9 -0.100% -3.245% 0.474% -0.926% 0.086% -0.825% 0.414% 3.989%
10 1.256% -1.989% 1.770% 0.844% 1.136% 0.312% 1.683% 5.672%
11 0.973% -1.016% 0.248% 1.092% 1.350% 1.662% 0.672% 6.344%
12 1.370% 0.354% -0.689% 0.402% 1.071% 2.733% -0.310% 6.034%
13 0.122% 0.477% -0.362% 0.040% 0.742% 3.475% -0.726% 5.308%
14 -2.175% -1.698% 3.997% 4.037% -1.008% 2.467% 3.174% 8.482%
15 0.361% -1.337% 0.110% 4.148% -0.224% 2.243% -0.668% 7.814%
16 -1.976% -3.314% 0.808% 4.955% -1.412% 0.831% 0.794% 8.608%
17 0.728% -2.585% 1.386% 6.341% -0.002% 0.829% 1.559% 10.167%
18 0.089% -2.497% -1.171% 5.170% 0.939% 1.769% -0.086% 10.082%
19 -2.143% -4.640% -0.351% 4.819% -1.837% -0.069% 0.121% 10.202%
20 1.574% -3.066% -2.378% 2.441% 1.938% 1.869% -2.207% 7.996%
42
(Table 10 continued)
Fama-French Model (Third Closest Contract, F3) Fama-French Model (Twelfth Closest Contract, F12)
Event SPR Increase SPR Decrease SPR Increase SPR Decrease
Day AR CAR AR CAR AR CAR AR CAR
-20 0.872% 0.872% 0.446% 0.446% 0.732% 0.732% 0.835% 0.835%
-19 -0.372% 0.500% 0.840% 1.286% -0.108% 0.625% 0.623% 1.458%
-18 0.825% 1.325% -0.698% 0.589% 0.967% 1.592% -0.814% 0.644%
-17 1.227% 2.552% -0.267% 0.322% 0.896% 2.488% -0.098% 0.546%
-16 -0.602% 1.950% 0.551% 0.874% -0.967% 1.521% 0.242% 0.788%
-15 -0.164% 1.786% 1.564% 2.437% -0.191% 1.329% 1.207% 1.995%
-14 1.003% 2.789% -0.699% 1.738% 0.942% 2.271% -0.662% 1.333%
-13 -0.499% 2.290% 0.445% 2.184% -0.485% 1.786% 0.171% 1.504%
-12 -0.322% 1.968% 0.614% 2.798% -0.341% 1.445% 0.341% 1.845%
-11 0.693% 2.661% -0.469% 2.329% 0.628% 2.072% -1.086% 0.759%
-10 -0.170% 2.490% -2.087% 0.241% -0.451% 1.621% -1.486% -0.727%
-9 -1.143% 1.347% -0.164% 0.078% -0.976% 0.645% -0.103% -0.831%
-8 -0.040% 1.308% 0.236% 0.314% -0.075% 0.570% 0.621% -0.209%
-7 -0.965% 0.343% -1.155% -0.841% -1.162% -0.592% -1.013% -1.222%
-6 -0.633% -0.290% 0.987% 0.146% -0.860% -1.451% 0.619% -0.603%
-5 -0.812% -1.103% 0.721% 0.867% -0.770% -2.222% -0.427% -1.029%
-4 1.229% 0.126% -0.145% 0.722% 0.840% -1.382% -0.032% -1.062%
-3 -0.909% -0.782% 0.943% 1.665% -0.871% -2.253% 1.052% -0.009%
-2 0.086% -0.696% 0.573% 2.238% 0.493% -1.760% 0.242% 0.233%
-1 -0.112% -0.808% 0.870% 3.109% 0.179% -1.581% 0.338% 0.571%
0 -1.056% -1.864% -3.219% -0.110% -0.719% -2.300% -1.148% -0.577%
1 -2.382% -4.246% -1.102% -1.212% -1.135% -3.435% -0.424% -1.001%
2 1.759% -2.487% 0.920% -0.292% 1.365% -2.070% 0.443% -0.558%
3 0.219% -2.268% 0.774% 0.483% 0.001% -2.069% 0.502% -0.056%
4 -0.068% -2.336% 0.164% 0.646% -0.268% -2.337% -0.331% -0.388%
5 1.097% -1.238% 0.582% 1.228% 0.659% -1.678% 0.405% 0.017%
6 -0.580% -1.819% 0.088% 1.316% -0.371% -2.049% -0.174% -0.157%
7 0.685% -1.133% 0.488% 1.805% 0.781% -1.268% 0.517% 0.360%
8 -0.292% -1.426% 1.073% 2.877% 0.152% -1.115% 0.773% 1.132%
9 0.024% -1.402% 0.338% 3.216% 0.150% -0.965% -0.005% 1.127%
10 0.738% -0.664% 1.416% 4.632% 0.433% -0.532% 0.911% 2.038%
11 1.310% 0.647% 0.479% 5.110% 1.094% 0.562% 0.062% 2.100%
12 0.892% 1.538% -0.221% 4.889% 0.616% 1.178% -0.301% 1.798%
13 0.621% 2.159% -0.695% 4.194% 0.516% 1.694% -0.535% 1.263%
14 -0.810% 1.349% 2.457% 6.651% -0.591% 1.103% 1.878% 3.142%
15 0.035% 1.384% -0.863% 5.788% 0.411% 1.513% -0.835% 2.306%
16 -1.344% 0.040% 0.419% 6.207% -0.958% 0.555% 0.347% 2.653%
17 0.071% 0.111% 1.339% 7.546% 0.117% 0.672% 0.968% 3.620%
18 0.845% 0.955% 0.011% 7.557% 0.911% 1.583% -0.094% 3.526%
19 -1.270% -0.314% -0.091% 7.467% -0.771% 0.812% -0.323% 3.203%
20 1.867% 1.552% -1.810% 5.657% 1.636% 2.448% -1.297% 1.906%
43
... The important advantage of using the event study method is that we can construct the measurement of Russia-Saudi Arabia oil price war's economic impact using the returns of crude oil futures prices and spot prices observed over a short time span (MacKinlay, 1997). Thus, the event study method has been widely used to measure how markets respond to events in the short term, such as energies (e.g., Demirer and Kutan, 2010;Karali et al., 2019;Ramiah et al., 2019;Sabet and Heaney, 2016); equities (e.g., Aitken et al., 1998;Lane and Jacobson, 1995;Lyon et al., 1999;Yang et al., 2015); and marketing (e.g., Chen et al., 2012;Swaminathan and Moorman, 2009;Wiles and Danielova, 2009). As for the dataset, we use the daily futures and spot prices of three major crude oil markets (i.e., West Texas Intermediate (WTI), European Brent, and Oman) to analyze how two events, the outbreak and truce (end) of the price war, influenced global crude oil returns. ...
... The event study approach is a widely applied analytical tool to investigate the effects of various events on markets or corporations (Demirer and Kutan, 2010). It is based on the efficient market hypothesis (EMH), which assumes that futures prices can incorporate new information quickly, since traders continually re-evaluate the value of futures contracts (Pozo and Schroeder, 2016). ...
... Generally, the market model uses an overall market index as a proxy to predict normal returns, then evaluate deviations between realized returns and normal returns; this method is explicit and relatively easy to use. Hence, it is widely used as a benchmark to assess abnormal returns in financial economics (e.g., Black and Kim, 2012;Brown and Warner, 1985;Demirer and Kutan, 2010;Draper, 1984;O'hara and Shaw, 1990;Pozo and Schroeder, 2016;Zhu et al., 2020). For instance, in terms of commodity markets, Demirer and Kutan (2010) use an event study with a standard market model to measure how crude oil spot and futures prices react to OPEC and U.S. Strategic Petroleum Reserve (SPR) announcements. ...
Article
The COVID-19 pandemic damaged crude oil markets and amplified the consequences of uncertainty stemming from the Russia-Saudi Arabia oil price war in March–April of 2020. We investigate the impacts of the oil price war on global crude oil markets. By doing so, we use the daily futures and spot prices in three major crude oil markets—West Texas Intermediate, European Brent, and Oman—to perform a systematic analysis of the impacts of the oil price war on them. The event study method, a well-established analytical tool to measure the impacts of a given event on markets, is used in this study. The results indicate that information leakage plays an important role in the impacts of the price war. The outbreak of and truce following the price war have asymmetrical impacts on the markets; negative impacts generated by information leakage during the outbreak are generally more durable than the positive ones it generated during the truce. Furthermore, the magnitude of the impacts on futures markets is negatively correlated with the time-to-maturity of futures. Finally, negative crude oil prices affect West Texas Intermediate crude oil markets the most. Our findings generally show that market participants could perceive and assimilate market changes and adjust their expectations, which restrained the impacts that should have occurred within the oil price war.
... An event study traces abnormal effects to determine the duration of a suspected market disturbance. Event studies of oil price shocks [142,143], for instance, have evaluated OPEC announcements [144,145] and storms [146]. Conversely, temporal clustering uses economic anomalies to extract events for further examination amid the flow of financial history. ...
... Extensions of this work can critical moments identified through unsupervised machine learning with event studies. In addition to OPEC announcements [144,145], the public disclosure of decisions affecting major agricultural markets and the resolution of global trade disputes over agriculture can serve as bases for comparative analysis. ...
Article
Full-text available
The identification of critical periods and business cycles contributes significantly to the analysis of financial markets and the macroeconomy. Financialization and cointegration place a premium on the accurate recognition of time-varying volatility in commodity markets, especially those for crude oil and refined fuels. This article seeks to identify critical periods in the trading of energy-related commodities as a step toward understanding the temporal dynamics of those markets. This article proposes a novel application of unsupervised machine learning. A suite of clustering methods, applied to conditional volatility forecasts by trading days and individual assets or asset classes, can identify critical periods in energy-related commodity markets. Unsupervised machine learning achieves this task without rules-based or subjective definitions of crises. Five clustering methods—affinity propagation, mean-shift, spectral, k-means, and hierarchical agglomerative clustering—can identify anomalous periods in commodities trading. These methods identified the financial crisis of 2008–2009 and the initial stages of the COVID-19 pandemic. Applied to four energy-related markets—Brent, West Texas intermediate, gasoil, and gasoline—the same methods identified additional periods connected to events such as the September 11 terrorist attacks and the 2003 Persian Gulf war. t-distributed stochastic neighbor embedding facilitates the visualization of trading regimes. Temporal clustering of conditional volatility forecasts reveals unusual financial properties that distinguish the trading of energy-related commodities during critical periods from trading during normal periods and from trade in other commodities in all periods. Whereas critical periods for all commodities appear to coincide with broader disruptions in demand for energy, critical periods unique to crude oil and refined fuels appear to arise from acute disruptions in supply. Extensions of these methods include the definition of bull and bear markets and the identification of recessions and recoveries in the real economy.
... Figure 1 shows that, the event study method usually denotes = 0 as the date of the event, ∈ [ , ] as the event window, and ∈ [ , ] and ∈ [ , ] as the estimation and post-event windows, respectively. Following the works [40,41], we select [−15, +15] as the event window to test the effect of exogenous event shocks on the risk spillover effect. The estimation window is set to [−45, −15] in this study because the estimation window is set to effectively avoid crossover effects of events and changes in the data structure. ...
Article
Full-text available
Based on the DY spillover index model, we explore the static and dynamic risk spillover relationships between the Chinese carbon and stock markets from the perspective of the entire market and different industry levels. Furthermore, we examine the impact of diverse types of exogenous events on the risk spillover effects. The empirical results of the sectoral stock market show that the carbon market is the primary risk taker, and the risk spillover to the carbon market is mainly from high-carbon-emitting industries, such as the oil and electricity industries. However, the risk spillover relationship will be reversed under the shocks from exogenous events. The shocks from different types of exogenous events enhance the risk spillover from the carbon market to the stock market, specifically to the oil sector. The Sino–U.S. trade war and the COVID-19 outbreak are more impactful than government policies. These findings help investors to understand the risk conduct patterns among different financial sub-markets, and have implications for regulators to strengthen market risk management.
... While statistical intervention analyses are common tools in studying the air quality (Fassò, 2013;Grange and Carslaw, 2019;Petetin et al., 2020) and the impact of pollution mitigation policies, ES are only recently receiving attention in pollution-related fields, such as energy and oil commodity markets. For example, Demirer and Kutan (2010) Zha et al. (2018) the authors aim at assessing the impact of refined oil price adjustments to control air pollution in China between 2014-2015. In addition, ES methods have recently received great attention in climate policy analysis. ...
Preprint
Full-text available
Event Studies (ES) are statistical tools that assess whether a particular event of interest has caused changes in the level of one or more relevant time series. We are interested in ES applied to multivariate time series characterized by high spatial (cross-sectional) and temporal dependence. We pursue two goals. First, we propose to extend the existing taxonomy on ES, mainly deriving from the financial field, by generalizing the underlying statistical concepts and then adapting them to the time series analysis of airborne pollutant concentrations. Second, we address the spatial cross-sectional dependence by adopting a twofold adjustment. Initially, we use a linear mixed spatio-temporal regression model (HDGM) to estimate the relationship between the response variable and a set of exogenous factors, while accounting for the spatio-temporal dynamics of the observations. Later, we apply a set of sixteen ES test statistics, both parametric and nonparametric, some of which directly adjusted for cross-sectional dependence. We apply ES to evaluate the impact on NO2 concentrations generated by the lockdown restrictions adopted in the Lombardy region (Italy) during the COVID-19 pandemic in 2020. The HDGM model distinctly reveals the level shift caused by the event of interest, while reducing the volatility and isolating the spatial dependence of the data. Moreover, all the test statistics unanimously suggest that the lockdown restrictions generated significant reductions in the average NO2 concentrations.
... Analysis on some cases in his study suggests that the fluctuation of abnormal returns affected by OPEC announcements can be ±5%. A study was conducted by Demirer and Kutan (2010), on the US Strategic Petroleum Reserve (SPR) and OPEC's announcements released between 1983 to 2008 on the spot and future oil prices. The results indicate that following the announcement, an abnormal return of the related markets witnessed apparent fluctuations. ...
Article
Full-text available
This paper examines the impact of the sentiments of OPEC news on stock market prices of public listed oil and gas companies in Bursa Malaysia. We used data of stock market prices from randomly selected oil and gas companies for the period of 2012 to 2017. For the methodology, we first established a supervised machine learning algorithm-based news classifier to classify the OPEC news following its sentiments. We developed a financial news sentiment classifier by combining machine learning algorithms and lexicon-based labelling methods. We then applied the event study method to investigate how stock market prices react to OPEC news’ sentiment. The results showed a negative correlation between OPEC news sentiment and stock market prices of oil and gas companies during the event window based on each OPEC news release date. The results further showed that the stock market prices do not react to OPEC news sentiment on event day. These findings should provide some guides to stock investors on the movement of the selected stock market prices of energy sector companies during the event window period.
... See, for example,Charles and Darné (2017),Larsson and Nossman (2011),Lee and Mykland (2008), andSalisu and Fasanya (2013). 2 SeeDemirer and Kutan (2010) andLin and Tamvakis (2010). ...
Article
Full-text available
This paper compares Generalized Autoregressive Score (GAS) models and GARCH-type models on their forecasting abilities for crude oil and natural gas spot and futures returns from developing and developed markets over multiple horizons. The out-of-sample forecasting results based on two loss functions and the Diebold–Mariano predictive accuracy test for multiple models show that the GAS framework outperforms GARCH and EGARCH models, particularly for crude oil assets. For natural gas, no specific model retains an advantage over the other two models as the predictive accuracy changes over forecasting horizons and varies across markets. Meanwhile, the GAS model performs well in both developed and developing markets. The Cumulated Sum of Squared Forecast Error Differential (CSSFED) graphically monitors the evolution of the relative forecasting performance of different models, and shows that the superiority of GARCH is vulnerable to extraordinary event shocks. Over the short-term forecasting (less than or equal to one month ahead), the GAS framework shows a prominent advantage over GARCH and EGARCH models for crude oil assets.
Article
With the acceleration of global energy transition and financialization, intense climate policy uncertainty and financial speculation have significant impacts on the global energy market. This paper uses TVP-VAR-SV models to analyze the nonlinear effects of climate policy uncertainty (CPU), financial speculation, economic activity, and US dollar exchange rate on global prices of crude oil and natural gas respectively, and then compare the time-varying response of oil prices and gas prices to six representative CPU peaks. The results show that responses of energy prices to various shocks have significant nonlinear effects: the time-varying effect of CPU on energy prices from positive to negative over time is significant, and financial speculation has the opposite effects on oil and gas prices. The effect from economic activity is mainly positive, while the effects of US dollar exchange are negative and stable. These results provide important implications for policymakers and investors dealing with high levels of climate policy uncertainty, financial speculation, and global economic activity.
Article
Subsidies are crutial for the development of China's New Energy Vehicle (NEV) market. With the country's ambitious goal to achieve carbon neutrality in 2060, NEV is playing a growing role in decarbonising the transport sector. This paper empirically evaluates the effectiveness of three main forms of subsidies: fiscal policy, preferential tax, and government procurement. Specifically, we pay attention to financing issues and use an event study approach to investigate how the capital market responds to these policies. Results show that both preferential tax and government procurement policies have significant positive impacts on NEV stock prices. Relative to lithium-ion battery and NEV manufacturers, charging pile companies benefit less from preferential tax or government procurement. For fiscal subsidy policies, their impacts on NEV stock prices are ambiguous, depending largely on the specific policy contents. An increase in targeted subsidies for those technology-oriented NEV firms has a positive impact on the stock market. However, subsidies falling short of expectations due to frequent policy changes and the removal of local government subsidies are associated with significant negative effects. At this critical stage when the Chinese government is re-designing subsidy policies on NEVs, special attention should be paid to setting a transparent schedule and reducing the potential negative effect caused to the stock market.
Article
Sparse and short news headlines can be arbitrary, noisy, and ambiguous, making it difficult for classic topic model LDA (latent Dirichlet allocation) designed for accommodating long text to discover knowledge from them. Nonetheless, some of the existing research about text-based crude oil forecasting employs LDA to explore topics from news headlines, resulting in a mismatch between the short text and the topic model and further affecting the forecasting performance. Exploiting advanced and appropriate methods to construct high-quality features from news headlines becomes crucial in crude oil forecasting. This paper introduces two novel indicators of topic and sentiment for the short and sparse text data to tackle this issue. Empirical experiments show that AdaBoost.RT with our proposed text indicators, with a more comprehensive view and characterization of the short and sparse text data, outperforms the other benchmarks. Another significant merit is that our method also yields good forecasting performance when applied to other futures commodities.
Article
Event studies focus on the impact of particular types of firm-specific events on the prices of the affected firms' securities. In this paper, observed stock return data are employed to examine various methodologies which are used 111 event studies to measure security price performance. Abnormal performance is Introduced into this data. We find that a simple methodology based on the market model performs well under a wide variety of conditions. In some situations, even simpler methods which do not explicitly adjust for marketwide factors or for risk perform no worse than the market model. We also show how misuse of any of the methodologies can result in false inferences about the presence of abnormal performance.
Article
Event study methodology has been one of the most frequently used tools in financial research in recent years. This study provides a review of the present state of knowledge and practice with respect to event study methodology. Many variations of this methodology are discussed, as well as special issues and applications. Recommendations for implementing an event study also are provided. [ABSTRACT FROM AUTHOR] Copyright of Quarterly Journal of Business & Economics is the property of College of Business Administration/Nebraska and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Article
This chapter presents the spectrum of experimental designs adopted by empiricists who employ event studies methods. In concert with the variety of detail, several investigations into empirical methods have concluded that minor variations have little impact on inferences. Researchers are left with discretion over the choice of estimation window, projection model, raw versus excess returns, forecast error versus event parameter and the form of hypothesis tests. Where these decisions are made ex ante, this discretion seems harmless although latitude can be manipulated, however unintentionally, to generate significant results in any specific application. If a researcher can choose an estimation window between 200 and 300 days, choose an event window between 1 and 5 days, select a projection model with 3 or 4 different types of explanatory variables, use raw or excess returns, pick parametric or nonparametric tests and exercise judgment over modeling how the event affects different firms, it is likely that something of interest will turn up in the data. Credibility is added to the findings of empirical investigations when the methods chosen can be defended on the basis of objective econometric or economic criteria, however minor the improvement in the estimation method on average.
Article
I analyze a simple test statistic for mean abnormal returns in the presence of stochastic volatility during both event and nonevent windows and in the presence of event-induced variance increases. Unlike previous tests, the parametric test evaluated here does not require that the volatility effect of the event be the same across all securities. Simulations show that the test exhibits nontrivial gains in power over previously developed parametric and nonparametric tests, and the true null hypothesis is rejected at appropriate levels. 2003 The Southern Finance Association and the Southwestern Finance Association.
Article
This paper reviews the literature on China's energy economy, focusing particularly on: (i) the relationship between energy consumption and economic growth; (ii) China's changing energy intensity; (iii) energy demand and energy-non-energy substitution; (iv) the emergence of energy markets in China; and (v) economic reforms in the energy industry. After reviewing the literature, the paper presents the main findings that some important issues remain unanswered, for example, what determines energy consumption behavior; the effects of substitution of and demand for energy; and technological change effects on energy intensity. Finally, the review suggests some topics worthy of future study.
Article
We consider a stylised model in which two cartel members bargain over the aggregate-production quota in a world of asymmetric information. We show that when the two cartel members are sufficiently different, the probability of agreement depends on both the current state of demand and initial production. Specifically, the probability of agreement is much lower when demand is low (and initial production is relatively high) than when demand is high (and initial production is relatively low). We also find that, regardless of the current demand state, the more extreme is initial production, the higher is the probability of agreement. Using an event study, where we take as events OPEC production quota announcements, we demonstrate empirically that the predictions of the model are borne out.