ArticlePDF Available

Gold Investment Incentives: An Empirical Identification of the Main Gold Price Determinants and Prognostication of Gold Price Future Trends

Authors:

Abstract and Figures

During the World War I, most of the countries stopped coin production and began converting paper money into gold. Various forms of exchange were later abolished during the "Great Depression" in 1929-1933. Later, gold lost the value of money in most of the economies worldwide. Multiple price rise of gold caused a real rise in the value of gold reserves and their potential ability to cover the balance of payment deficit. At the same time, it shows that gold still plays an important role in terms of monetary aspect. The aim of this study was to determine whether ARIMA models are suitable for determining the short-term volatility of gold prices. The calculations show that ARIMA model is suitable only for short-term gold price forecasts (max. 1 year). Thus, it is necessary to apply other models (multi-regression ones) that also can reveal the relationship between gold price and its determinants.
Content may be subject to copyright.
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
248
THE MAIN GOLD PRICE
DETERMINANTS AND THE
FORECAST OF GOLD PRICE FUTURE
TRENDS
Ligita Gaspareniene,
Lithuanian Institute of Agrarian
Economics
Vilnius, Lithuania
E-mail: ligita.gaspareniene@laei.lt
Rita Remeikiene,
Lithuanian Institute of Agrarian
Economics
Vilnius, Lithuania
E-mail: rita.remeikiene@laei.lt
Alius Sadeckas,
JCC “Ekskomisarų biuras“,
Vilnius, Lithuania
E-mail: alius@ebiuras.lt
Romualdas Ginevicius,
Vilnius Gediminas Technical
University, Vilnius, Lithuania
E-mail:
romualdas.ginevicius@vgtu.lt
Received: January, 2018
1st Revision: April, 2018
Accepted: June, 2018
DOI: 10.14254/2071-
789X.2018/11-3/15
ABSTRACT. During the World War I, most of the
countries stopped coin production and began converting
paper money into gold. Various forms of exchange were
later abolished during the "Great Depression" in 1929-
1933. Later, gold lost the value of money in most of the
economies worldwide. Multiple price rise of gold caused a
real rise in the value of gold reserves and their potential
ability to cover the balance of payment deficit. At the same
time, it shows that gold still plays an important role in
terms of monetary aspect. The aim of this study was to
determine whether ARIMA models are suitable for
determining the short-term volatility of gold prices. The
calculations show that ARIMA model is suitable only for
short-term gold price forecasts (max. 1 year). Thus, it is
necessary to apply other models (multi-regression ones)
that also can reveal the relationship between gold price and
its determinants.
JEL Classification
: D25
Keywords
: gold price, gold price variation, autoregressive model.
Introduction
Over the last few decades, global finance markets have gone through a number of
financial crises, the most devastating of which include Mexican peso crisis in 1994, Asian flu
in 1997, Russian crisis in 1998, Brazilian crisis in 1999, Argentinian crisis in 2001-2002, the
USA financial crisis in 2007 and Greek crisis in 2009. The periods of financial contagion
have evidently raised the risk of securities (Vychytilova, 2018; Mačí & Valentová
Gaspareniene, L., Remeikiene, R., Sadeckas, A., & Ginevicius, R. (2018). Gold
Investment Incentives: An Empirical Identification of the Main Gold Price
Determinants and Prognostication of Gold Price Future Trends. Economics and
Sociology, 11(3), 248-264. doi:10.14254/2071-789X.2018/11-3/15
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
249
Hovorková, 2017; Čulková et al., 2015; Vukovic et al., 2017) and returned the interest in gold
as in alternative financial instrument since gold has historically been treated as a standard of
high value. Of course, other alternative investments should be mentioned here too
(Jurevičienė, & Jakanovytė, 2015; Nuhiu et al., 2017; Mouselli et al., 2016; Śliwiński &
Łobza, 2017; Nawrocki, 2018). The series of the global financial crises has even augmented
the belief that gold can provide investment protection and serve as a perfect risk management
tool. As it was noted by (Baur, & McDermott, 2010; Ohanyan, & Androniceanu, 2017), the
main difference between gold and other types of assets lies in gold’s positive reaction to
unwanted shocks at financial markets. Gold value reached its historical heights in the 1980s,
when the global economies were facing the threat of stagflation brought about by the
petroleum crisis of the 1970s. Similar trends were observed during the 2007 US sub-prime
financial crisis which gathered its pace in September 2008, when gold price started soaring
(Baur, McDermott, 2010).
Leaning on the assumption that gold can be treated as a store of value, investors need
to know what role is played by gold while forming an investment portfolio, i.e., when gold is
attributed to one of the assets categories. Considering the inclusion of gold in an investment
portfolio is extremely important minding the fact that gold markets (like all other financial
markets) are characterised by volatility and speculations.
A growing interest in gold as an investment has prompted conducting this research
focused on gold investment, the links between gold and other financial instruments (e.g.,
securities) and the efficiency of gold as an instrument of financial risk management.
Markowitz’s (1952, 1959) studies laid the foundations for practical assessment of the benefits
gained from an investment portfolio diversification and proved that combination of a few
categories of assets may significantly reduce portfolio value fluctuations (Vukovic & Prosin,
2018). The importance of gold as of an investment was highlighted by (Jaffe, 1989; Michaud
et al., 2006; Conover et al., 2009; Riley, 2010; Baur, 2013; Bradfield, & Munro, 2016) and
many others. The specific role of gold in diversification of portfolio investment was analysed
by (Sherman, 1982; Adrangi et al., 2000; Smith, 2002; Liu, & Chou, 2003; Davidson et al.,
2003; Lucey, Tulley, 2006 a,b; Ibrahim, 2012; Makiel, 2015; Brycki, 2015; Bundrik, 2016;
Yu, H.-C., Lee, C.-J. & Shih, T.-L., 2016; Giannarakis, G., Partalidou, X, Zafeiriou, &
Sariannidis, N., 2016) and others. Nevertheless, different scientific studies often provide
contradictory results concerning gold investment incentives and an optimal share of gold in an
investment portfolio. This contradiction, in its turn, calls for more comprehensive research in
this area. This article is aimed at empirical identification of the main gold price determinants
and forecast of gold price future trends. The defined aim was detailed into the following
objectives: 1) to review the recent gold demand trends; 2) to identify the theoretical gold
investment incentives and optimal shares of gold in low-, medium- and high-risk investment
portfolios; 3) to select and substantiate the methodology for this research; 4) modelling the
autoregressive process, basing on time series observations for the previous years, to identify
the main gold price determinants and forecast the gold price future trends.
The methods applied in the research include systematic and comparative analysis of
scientific literature, ARMA/ARIMA models, which are flexible forecast models based on the
employment of historical information. ARMA/ARIMA models are composed of the
autoregressive (AR) process, moving average (MA) process and integration (I) process.
1. Review of the recent gold demand trends
Although in modern economies gold has stopped being used as a tool for daily
settlements, its role in the global economy is still significant. The data in the balance sheets
provided by central banks and other financial institutions, such as the IMF, show that the
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
250
above-mentioned institutions accumulate gold reserves and generate nearly one-fifth of the
global demand for gold (Balarie, 2017). According to Ghosh et al. (2004), the demand for
gold is characterised by two structural components:
1) the need for the direct use of gold (for instance, gold is directly used for jewellery,
medal minting, electronic industry, dentistry, etc.);
2) the need for gold as an asset (governments, fund managers, individual investors buy
and hold gold as an investment).
The trends of the demand for gold for different purposes have been depicted in
Table 1.
Table 1. The trends of the demand for gold in tones and million US dollars (with reference to
the data for 2015)
Demand
type
The demand for gold, tonnes
The demand for gold, monetary value
(mln. US dol.)
2014/Q4
2015/Q4
5-year
average
Annual
change
2014/Q4
2015/Q4
5-year
average
Annual
change
Jewellery
677.4
671.4
585.0
-1%
26166.5
23885.4
26659.9
-9%
Technologies
90.3
84.5
93.3
-7%
3489.8
3004.8
4314.5
-14%
Investment
169.3
194.6
298.9
15%
6540.4
6922.6
14308.8
6%
Gold
bullions and
coins
260.9
263.5
328.8
1%
10075.9
9372.7
15343.6
-7%
ETFs and
similar
financial
products
-91.5
-68.9
-29.8
-
-3535.5
-2450.0
-1034.8
-
The demand
generated by
central banks
and other
financial
institutions
133.9
167.2
133.1
25%
5173.5
5948.7
6115.2
15%
Total
demand
1071.0
1117.7
1110.3
4%
41370.2
39761.5
51398.4
-4%
Source: compiled by the authors with reference to the data announced by the World Gold Council, 2016, p. 2.
As it can be seen in Table 1, the total demand for gold in the fourth quarter of 2014
amounted to 1071 tonne (or 41370.2 mln. US dollars), while in the fourth quarter of 2015 it
increased up to 1117.7 tonnes (or 39761.5 mln. US dollars; the lower monetary value shows
that the price of gold decreased over the period under consideration). Jewellery industries
generate the largest share of the total demand for gold; a smaller but also significant share of
the total demand is generated by investors in gold bullions and coins. With reference to the
data announced in the World Gold Council’s report for 2016 (the World Gold Council, 2017),
in comparison to 2015, in 2016 a full-year jewellery demand suffered a sharp 15 percent
decline, while investment demand increased up to 70 percent, reaching its highest value
since 2012.
Summarising, the total demand for gold is gradually rising with 2016 full-year demand
reaching its 3-year height of 4.308 tonnes. The annual investment inflows in ETFs and similar
financial products are also growing, which is confirmed by ETFs’ second highest rate on the
statistical record. The declines in the demand for gold generated by jewellery industries and
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
251
central banks offset the growth in the demand for ETFs, but the annual demand for gold
bullions and coins remains broadly stable.
2. The incentives to invest in gold and an optimal share of gold in low-, medium- and
high-risk investment portfolios
As it was noted by Balarie (2017), gold is attractive for modern investors since gold
prices do not correlate to stock, bond or real estate prices. Some recent scientific studies
(Barber, & Odean, 2000; Coval, & Moskowitz, 2001; Mouna, & Jarboui, 2015; Balarie, 2017)
disclosed two behavioural patterns which contradict the assumptions of the classical portfolio
theory: 1) modern investors fail to diversify their portfolios;
2) asset classes in investment portfolios are unreasonably interrelated.
These findings show that modern investors are inclined to form poorly diversified
investment portfolios from a limited number of asset classes (Barber, Odean, 2000). Rather
than following rational motives, many investors are likely to select particular asset classes that
match their professional area or are close to their living place (Coval, & Moskowitz, 2001).
Nevertheless, a rational and well-informed investor should diversify his/her investment
portfolio regardless of the level of risk assumed (Mouna, & Jarboui, 2015; Mentel, &
Brożyna, 2015).
The literature in economics and finance proposes three main motives to include gold
in an investment portfolio for its diversification:
- gold may serve as a hedge against inflation (money devaluation);
- gold may serve as a hedge against exchange rate risks;
- gold may serve as a measure to reduce the overall risk of a portfolio.
The idea that gold may serve as a hedge against inflation is based on the observation
that it is wise to hold physical assets (e.g., real estate, merchandise) during inflationary
periods. Unlike other kinds of merchandise, gold is durable, transportable, universally
acceptable and easily authenticated (Worthington, & Pahlavani, 2007). Historical evidence
shows that the rise in an inflation rate causes gold prices to increase (Balarie, 2017). For
instance, when in January 2006 the price of gold exceeded $562 per ounce, the average
monthly change amounted to 0.20225 percent over the period 1875-2006. It was higher than
the average monthly change in the US consumer price index (the latter amounted to 0.2022
percent) over the same period (Worthington, & Pahlavani, 2007). Nevertheless, as it was
noted by Aggarwal (1992), gold serves as an inflation hedge only in the long run, while its
efficiency in the short and medium run is debatable. Eryiğit (2017) finds precious metals and
energy to have large effect on the gold price being gold price main determinants.
The issue of exchange risk management is extremely topical nowadays, when stock
return is difficult to forecast, and bond return is comparatively low. In fact, even investors
who refrain from direct international investment are facing the threat that the strong
correlation between the local asset value and inflation as well as between the local asset value
and capital outflows may negatively affect local exchange rates. When the differences in
interest rates in developing and developed economies are significant, exchange rate
management costs can substantially reduce the return on investment. Historically,
employment of other currencies has never been an efficient hedge again exchange rate risk.
What is more, the marginal reduction in an exchange rate, which determines a substantial
reduction in the potential return on investment, is not always attractive for investors. A
potential solution to this situation is employment of gold. As gold is nobody’s debenture, it
can help investors manage foreign asset risks, especially in the countries where exchange
rates are highly volatile, and interest rates are structurally high. With reference to the report of
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
252
the World Gold Council (2013), gold prices positively correlate to the growth pace in
emerging markets, and negatively correlate to the value of the US dollar. At the same time,
the cost of investment in gold remain comparatively low.
An increasing role of emerging economies in the global gold market (in particular,
over the last 12 years) is another rationale to use gold as a hedge against exchange rate risks.
In fact, the correlation between gold and stock prices in emerging markets is much higher
than the correlation between gold and stock prices in developed markets. For instance, over
the period 2001-2012, the correlation between gold and stock prices in emerging markets was
equal to 0.28, while the correlation between gold and stock prices in developed markets was
equal to 0.11 (The World Gold Council, 2013).
Finally, inclusion of gold in an investment portfolio should be considered for the
reason that gold can serve as a hedge against the overall risk of a portfolio. Menon (2015)
provides some arguments why gold can be treated as a highest-class investment:
1) the demand for gold in China, India and other emerging markets is soaring;
2) the growth in the global gold supply is limited since the world has already exploited
its most efficient gold mines;
3) gold production costs are expected to rise, which should cause the price of gold to
grow. Gold price is highly interlinked with the inflation expectations driving future monetary
development in an economy (Belke, 2017). According to Dar and Maitra (2017), the fact that
gold retains its value during the periods of economic crises, when the value of other assets is
dramatically decreasing, has been historically proven. In recent years, the values of different
assets have started showing stronger correlations. The results of some scientific studies reveal
that gold negatively correlates to stock (Baur, & Lucey, 2010; Baur, & McDermott, 2010 and
others). Although the US financial markets were coming through a sudden decline from July
2007 to March 2009, gold prices at the same time increased by nearly 42 percent (Baur, &
McDermott, 2010). Wang et al. (2016) note that extreme risks are much faster transmitted
after a crisis than before it, and this feature is linked to the efficiency of gold as a “safe
heaven”. The importance of gold in global financial systems is determined by the interest in
gold shown by both institutional and individual investors who treat gold as an alternative asset
(Shahbaz et al., 2014). The studies focused on the links between gold and different
macroeconomic and financial indicators reveal that gold can offset the fluctuations of several
macroeconomic and financial indicators, in particular, general prices, stock prices, crude oil
prices and exchange rates (Dar, & Maitra, 2017).
Along with the incentives to include gold in an investment portfolio, it is purposeful to
research which share of gold could be considered optimal in a low-, medium- and high-risk
investment portfolio. Leaning on the resampled efficiency approach, Michaud et al. (2006)
estimated that gold should compose 1-2 percent of low-risk investment portfolios, and 2-4
percent of risk-balanced investment portfolios. Having research the US market, the authors
(Michaud et al., 2006) found that over the last 32 years gold had been an important structural
component of risk-balanced investment portfolios, but the same could not be said about high-
risk investment portfolios. Leaning on the portfolio optimisation approach, Artigas (2010)
found that gold could mitigate the fluctuations in a portfolio’s value without sacrificing a part
of expected returns. The author (Artigas, 2010) concluded that inclusion of gold not only
provided a better risk-return balance, but also helped to reduce a potential loss on an
investment. A special attention was drawn to the finding that gold was able to reduce the level
of financial risks, i.e., the value of VaR coefficient. Artigas’s (2010) study disclosed that even
a relatively small share of gold (about 2.5-9 percent) in an investment portfolio might
significantly increase a risk-balanced return and diminish a portfolio’s value-at-risk
coefficient (VaR) by 0.1-18.5 percent.
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
253
The study carried out by Baur (2013) revealed that an optimal share of gold in an
investment portfolio should amount to 0-15 percent of a portfolio’s efficiency frontier. Gold
should compose the maximum share (about 15 percent) of an investment portfolio if an
investor pursues a target (desired) return; it should amount to 7-11.3 percent of the total
structure of an investment portfolio if an investor pursues a high, but lower than a target
return; finally, gold should compose about 4-7 percent of an investment portfolio if an
investor pursues a comparatively low return (Baur, 2013).
It should not be overlooked that a stage of an economic cycle can also have a
significant impact on the rate of return on a portfolio. With reference to the methodologies
introduced by Munro and Silberman (2008), economic cycles can be divided into 4 stages
economic development, stagflation, revival and shrinking. Structural shares of gold in an
optimal investment portfolio in different stages of an economic cycle have been depicted in
Graph 1.
Graph 1. Structural shares of gold in an optimal investment portfolio in different stages of an
economic cycle when a target return is equal to СPI+5%
Source: compiled by the authors with reference to Bradfield and Munro, 2016, p. 184.
With consideration of the impact of inflation on the real return on a portfolio during
the entire economic cycle, CPI+5% (consumer price index plus 5 percent) was selected as an
investor’s target return. The data in Graph 1 illustrate that the largest structural share (27
percent) of gold in an optimal investment portfolio is inherent to the stage of economic
revival, while the smallest share of gold (about 2 percent) should be included in an optimal
investment portfolio during the periods of economic stagflation. In the stage of economic
shrinking, the structural share of gold in an optimal investment portfolio should be equal to
the average structural share of gold in a portfolio over the entire economic cycle (i.e. it should
amount to 4 percent of the total structure of a portfolio).
Summarising, although the demand for gold as an asset is traditionally linked to the
attitude that gold may serve as an efficient hedge against inflation, the results of previous
scientific studies proved the efficiency of this strategy only in the long run. Since gold is
nobody’s debenture, it can help investors manage foreign asset risks at low costs, especially in
the economies where exchange rates are highly volatile, and interest rates are structurally
high. Gold is also considered to be a hedge against the overall risk of a portfolio since it
retains its value during the periods of economic crises when the values of other assets are
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
254
dramatically decreasing; what is more, gold ensures diversification of investment and
provides protection against so-called “tail risk”, i.e., against the risk that a portfolio return
may deviate from its average within the amplitude higher than three regular values of standard
deviation. The negative correlation between gold and exchange rates in developed economies
(apart from the US and the US dollar) creates the basis for worrying about particular
weaknesses inherent to the global monetary system. Therefore, gold is employed as a measure
that can protect investment from extreme changes in the global monetary system. Even if gold
cannot be referred to as a perfect substitute to the currencies of developing economies,
inclusion of gold in an investment portfolio allows to balance investment risks and earn
higher portfolio returns. With reference to Michaud et al. (2006), gold should compose 1-2
percent of low-risk investment portfolios, and 2-4 percent of well-balanced investment
portfolios. Inclusion of gold is not so important in high-risk investment portfolios since
investors who assume high risks for high return are not inclined to employ any hedges
because it would mean the decrease in expected returns. Over the periods of economic
development, stagflation and shrinking gold should compose 4-5 percent of the total structure
of an investment portfolio, while over the periods of economic revival the structural share of
gold should increase to 27 percent. In general, gold may provide some financial benefits if
one considers that gold’s main advantage lies not in its capability to raise portfolio returns,
but its ability to reduce portfolio risks.
3. Methodological approach
When comparing the results of different scientific studies focused on the benefits of
investment in gold, correlations of the price of gold to the prices of other assets, and
estimation of the share of gold in an optimal investment portfolio, contradictions of the results
can be observed. This is partly determined by employment of different research
methodologies. The methodologies commonly employed for the analysis of the investment in
gold can be attributed to three main categories:
1) the methodologies that allow to identify the links between gold price variations
and the changes in the main macroeconomic indicators (exchange rates, interest rates, income
level, political shocks, etc.) (Ariovich, 1983; Dooley et al., 1995; Sherman, 1982, 1983, 1986;
Sjaastad, & Scacciallani, 1996 and others);
2) the methodologies focused on the speculations and/or rational assessment of
the price of gold (Baker, & Van Tassel, 1985; Diba, & Grossman, 1984; Koutsoyiannis, 1983;
Pindyck, 1993 and others);
3) the methodologies that reveal (in)efficiency of gold as a hedge against
undesirable economic changes (e.g., inflation, the changes in exchange rates, the changes in
stock prices, etc.) in the short and long run (Chappell, & Dowd, 1997; Kolluri, 1981; Laurent,
1994; Mahdavi, & Zhou, 1997; Moore, 1990; Ghosh et al., 2000 and others).
The comprehensive literature analysis allowed to select two methods that enabled to
forecast the changes in the price of gold and identify the most significant determinants of gold
price variations over the period 1968-2015.
Time series analysis refers to the analysis of the data series extracted from various
databases (e.g., trading economics). In this research, employment of the paid data ensured
obtainment of the sufficiently long time data series. The model of time data series comprised
the following stages:
1. Recognition, i.e., preliminary selection of the model for the analysis. In this stage,
the initial data graph was presented, the nature of the trend was identified, and the effect of
seasonality (if any) was established.
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
255
2. Interim evaluation, i.e., setting of the model parameters. If the selected parameters
failed to describe time series with appropriate accuracy, the research was returned to the stage
of recognition.
3. Final evaluation, i.e., assessment of applicability of the model by particular criteria.
The method of time data series allowed to:
develop the structure of ARMA/ARIMA forecasting model;
ensure stationary;
assess the parameters of the selected model;
verify applicability of the model;
create forecasts by employing adequate models.
ARMA/ARIMA forecasting models are popular for their flexibility. They are based on
employment of historical information. ARMA/ARIMA forecasting models comprise an
autoregressive (AR) process, a moving average (MA) process and an integration (I) process.
The autoregressive process explains time series observations with consideration of historical
observations. The autoregressive equation, which describes the value of variable yt, is
expressed as follows:
, (1)
where:
yt time series observations;
a1; : : : ; ap parameters of the autoregressive model describing the dependence of each time
series value on its historical values;
"t random error;
p sequence of the autoregressive process.
The equation of the moving average process describing value yt is expressed as
follows:
(2)
Stationary process Yt is referred to as ARMA (p, q) if it satisfies the equation:
(3)
ARIMA (the AutoRegressive Integrated Moving Average) is an autoregressive integral
method of moving averages, which is often employed for time series analysis. It combines the
methods of autoregression, differentiation and moving averages. All structural components in
this model are based on the concept of random noise (inexplicable dispersal) which distorts the
systemic component of a time series. The structural components also react to the mode of the
noise. The most general ARIMA model covers all three structural parts, and can be expressed
as ARIMA (p, d, q), where p stands for an autoregression series, d for a differentiation series,
and q for a number of moving averages.
Construction of autoregressive time series models requires the series to be stationary. In
case the requirement of stationarity is not satisfied, different methods of transformation are
employed. Differentiation is one of the main methods of transformation. While diversifying
any time series in this research, the changes in its informational values were identified, and the
series was transformed into a stationary form:
Yt(1−a1L−a2L2 −...−apLp) = εt, (4)
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
256
Variable L in equation (4) stands for a lag operator. The feature of a lag operator is expressed
as LiYt = Yt−i.
For verification of the stationarity of a time series and development of a differentiation
series, the extensive Dickey-Fuller (ADF) test was employed. The hypotheses H0 and H1
were verified: H0 the process is not stationary; H1 the process is stationary. For
verification of hypothesis H0, the observed probabilities of significance (p) were employed.
Almost all statistical software packages were able to estimate these values. H0 was accepted if
p 0; otherwise, H1 was accepted. Any non-stationary time series was transformed into a
stationary series by employing the procedure of diversification:
δyt = yt −yt−1, (5)
If the differences in the first differentiation series were non-stationary, differentiation
of the second series was employed, and so forth. The preliminary parameters of
ARMA/ARIMA model were selected by analysing ACF and PACF graphs. In MA(q) process,
the serial number at which ACF values still significantly varied from zero were selected. In
AR(p) process, the serial number at which PACF values still significantly varied from zero
were selected. After selection of the preliminary model parameter values p, d and q, the
adequacy of these values was verified. The final choice of the model was assessed by AIC
(Akaike) criterion which was estimated by the formula:
AIC = −2logL + 2k, (6)
where:
k the number of the parameters in the model;
L Gauss-type probability function.
Applicability of the model developed for gold price forecast was verified by
employing the following forecast accuracy measures:
MAE medium absolute error;
MAPE medium absolute percentage error.
Further in the research, the gold price trends were analysed by applying
ARMA/ARIMA model. All the calculations were performed with R software.
ARMA/ARIMA models were found to be suitable and applicable only for detection of slight
data fluctuations in the short run. At the same time, the sudden changes in the data were
difficult to capture.
4. The results of the empirical research: gold price forecast and gold price determinants
The forecast of the gold price future trends was started from the analysis of the
primary data for the period December 1968 December 2015 (see Graph 2).
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
257
Graph 2. Primary statistical data on the prices of gold for the period 1968-2015
Source: compiled by the authors
As in this research we wanted to verify the accuracy of the model and check its
applicability for forecasting of the gold price future trends, first of all, we verified the
accuracy of the forecast for the period 1968-2015 (see Graph 3).
Graph 3. The continuous data on the price of gold till 2015
Source: compiled by the authors
The year 2015 was selected to obtain the data for identification of the errors in the
values under consideration (see Graph 4):
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
258
Graph 4. The dynamics of the price of gold in 2016
Source: compiled by the authors
The results of the research show that the primary data are not stationary. The initial
data graph obviously shows that the average price of gold is changing. This trend can be
explained as a result of the sensitivity of other investment instruments to the fluctuations of
the price of gold. Nevertheless, the investment in gold stock is considered to be the riskiest
since the prices of gold stock rise and fall faster than the price of gold. The investment in gold
funds and mutual funds is usually considered more effective than the investment in any
physical form of gold because investors do not face the problems of gold storage.
Further in the research, by employing ADF (the Augmented Dickey-Fuller test), we
verified the hypothesis that the data were non-stationary. The hypothesis on the data non-
stationarity was confirmed. For the development of a stationary time series, the process of
differentiation was employed: Ydt = Yt −Yt−1
After differentiation of the first series, we obtained a stationary time series with the
data fluctuating around the same value 0 (see Graph 5).
Graph 5. A time series
Source: compiled by the authors
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
259
The Dickey-Fuller test, repeatedly applied for verification of the hypothesis about data
non-stationarity, allowed to decline this hypothesis. Differentiation of the first series was
sufficient to obtain a stationary time series. At the same time, it was established that d = 1,
and it would definitely be included in the model. For estimation of the other parametrical
values (p and q), the graphs of autocorrelation (ACF) and partial autocorrelation (PACF)
functions were analysed. In graph ACF, the initial distinguishing values were attributed to
parameter q, while in graph PACF, the initial distinguishing values were attributed to
parameter p.
Next, the models ARIMA (p,d,q) with different parametrical values were compared.
For further research, two models with lowest values of AIC (Akaike) criterion, i.e., ARIMA
(0,1,1) with drift and ARIMA (1,1,1), were selected. Verification of the models covered
examination of their adequacy and compliance with the requirements for autoregressive
models. Gold price forecasts for 2015 from ARIMA (0,1,1) with drift have been depicted in
Graph 6.
Graph 6. Gold price forecasts for 2015 from ARIMA (0,1,1) with drift
Source: compiled by the authors
The results of the research show that ARIMA’s (0,1,1) with drift MAPE indicator,
estimated for the continuous data of 2015, amounts to 6.14 percent, i.e., it does not exceed 4
percent.
The results of ARIMA’s (1,1,1) adequacy verification have been presented in
Appendix 4. ARIMA’s (1,1,1) MAPE indicator, estimated for the continuous data of 2015,
amounts to 3.93 percent, i.e., it does not exceed 4 percent. Hence, ARIMA (1,1,1) is more
suitable for short-term gold price forecasting than ARIMA (0,1,1) with drift.
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
260
For more accurate forecasts, it would be purposeful to include a larger number of
variables, to conduct more comprehensive mathematical calculations and to employ a wider
variety of multiple regression models, which would allow to identify closer correlations
between particular determinants and the price of gold. The analysis of scientific literature has
revealed that the price of gold is affected by the fluctuations in the price of silver, the price of
platinum, inflation rates, etc. The results of the empirical analysis lead to the conclusion that
the model ARIMA (1,1,1) can be treated as suitable for forecasting of the gold price future
trends. This study examines the changes in the price of gold over the period January 1968 to
December 2015, inclusive. The initial data on the prices of gold over the period under
consideration were analysed and used for the gold price forecast.
Conclusion
1. Although the demand for gold as an asset is traditionally linked to the attitude that
gold may serve as an efficient hedge against inflation, the results of previous scientific studies
proved the efficiency of this strategy only in the long run. As gold is nobody’s debenture, it
can help investors manage foreign asset risks at low costs, especially in the economies where
exchange rates are highly volatile, and interest rates are structurally high. Gold is also
considered as a hedge against the overall risk of a portfolio since it retains its value during the
periods of economic crises when the value of other assets is dramatically decreasing; what is
more, gold ensures diversification of investment and provides protection against so-called
“tail risk”, i.e. against the risk that portfolio returns may deviate from their average within the
amplitude higher than three regular values of standard deviation.
2. When comparing the results of different scientific studies focused on the benefits of
investment in gold, identification of the links between gold price and other asset price
variations, and estimation of the share of gold in an optimal investment portfolio,
contradictions of the results can be observed. This is partly determined by employment of
different methodologies, which calls for development of a method suitable for forecasting of
the gold price future trends. The comprehensive analysis of scientific literature allowed to
select two methods that could help forecast gold price fluctuations and identify the most
significant determinants of the changes in the price of gold. These methods include time
series analysis and ARMA/ARIMA forecasting models. ARMA/ARIMA forecasting models
are popular for their flexibility. ARMA/ARIMA models comprise an autoregressive (AR)
process, a moving average (MA) process and an integration (I) process which are based on
employment of historical information. In order to conduct a consistent study and obtain
accurate and applicable results, the statistical data for the period 1968-2015 was employed for
this research.
3. The autoregressive process explains time series observations with consideration of
historical observations. ARIMA’s (1,1,1) MAPE indicator, estimated for the continuous data
of 2015, is equal to 3.93 percent, i.e. it does not exceed 4 percent and is lower than the same
indicator estimated for ARIMA (0,1,1) with drift, which proposes that ARIMA (1,1,1) can be
treated as a model suitable for forecasting of the gold price future trends. Nevertheless, it
should be noted that ARMA/ARIMA models are suitable only for identification of
comparatively insignificant data fluctuations in the short run, while sudden data fluctuations
are hard to detect. This feature inevitably reduces the efficiency of the models. For more
accurate forecasts, it would be purposeful to include a larger number of variables, to conduct
more comprehensive mathematical calculations and to employ a wider variety of multiple
regression models, which would allow to identify closer correlations between particular
determinants and gold price fluctuations.
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
261
4. The limitations of this empirical research are linked to the lack of a long dynamic
line for gold, silver and platinum prices, money supply and other variables (for the period
1968-2015). The study could have considered various factors determining the price of gold,
but due to the gaps in time lags, we eliminated them from our calculations. Further research
could address the impact of American, Chinese and European monetary policy measures on
the fluctuations of the price of gold.
References
Adrangi, B., Chatrath, A., & Christie, D. R. (2000). Price discovery in strategically-linked
markets: the case of the gold-silver spread. Applied Financial Economics, 10, 227-234.
Aggarwal, R. (1992). Gold Markets. In P. Newman, M. Milgate, & J. Eatwell (Eds.), The New
Palgrave Dictionary of Money and Finance (2nd ed., pp. 257-258). Basingstoke:
Macmillan.
Ariovich, G. (1983). The impact of political tension on the price of gold. Journal for Studies
in Economics and Econometrics, 16, 17-37.
Artigas, J. C. (2010). Rediscovering gold as an asset class. Journal of Indexes. Retrieved
from http://www.etf.com/publications/journalofindexes/joi-articles/8239-
rediscoveringgold-as-an-asset-class.html
Baker, S. A., & Van Tassel, R. C. (1985). Forecasting the price of gold: a fundamentalist
approach. Atlantic Economic Journal, 13, 43-51.
Balarie, E. (2017). Gold as a diversifying investment. Retrieved from:
http://www.investopedia.com/articles/basics/08/invest-in-gold.asp#ixzz4muhvTqIF
Barber, B., & Odean, T. (2000). Trading is hazardous to your wealth: the common stock
investment performance of individual investors. Journal of Finance, 55(55), 773-806.
Baur, D. G. (2013). Gold fundamental drivers and asset allocation. Retrieved from
https://www.tdr.wiwi.uni-due.de/fileadmin/fileupload/VWL-WIPO/WiWi-
Kolloquium/Baur_Paper.pdf
Baur, D. G., & Lucey, B. M. (2010). Is gold a hedge or a safe haven? An analysis of stocks,
bonds, and gold. The Financial Review, 45(2), 217-229.
Baur, D. G., & McDermott, T. K. (2010). Is gold a safe haven? International evidence.
Journal of Banking & Finance, 34(8), 1886-1898.
Belke, A. (2016). Global liquidity and strategies of exit from unconventional monetary
policies, Economic Research-Ekonomska Istraživanja, 29(1), 286-313, DOI:
10.1080/1331677X.2016.1168039
Bradfield, D., & Munro, B. (2016). The role of gold bullion in South African balanced
portfolios. South African Journal of Accounting Research, 30(2), 172-186.
https://doi.org/10.1080/10291954.2015.1105546
Brycki, C. (2015). Comment: whats the point of gold in an investment portfolio? Retrieved
from http://www.sbs.com.au/news/article/2015/11/11/comment-whats-point-gold-
investment-portfolio
Bundrik, H. M. (2016). How to include gold in your investment portfolio. Retrieved from
http://budgeting.thenest.com/include-gold-investment-portfolio-30434.html
Chappell, D., & Dowd, K. (1997). A simple model of the gold standard. Journal of Money,
Credit and Banking, 29, 94-105.
Ciobanu, A., & Androniceanu, A. (2018). Integrated human resources activities the solution
for performance improvement in Romanian public sector institutions. Management
Research and Practice, 10(3), September, 60-79.
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
262
Conover, C. M., Jensen, G. R., Johnson, R. R., & Mercer, J. M. (2009). Can precious metals
make your portfolio shine? The Journal of Investing, 18(1), 75-86.
http://dx.doi.org/10.3905/JOI.
Čulková, K., Csikósová, A., & Janošková, M. (2015). Development of risk payment index in
Slovakia comparing with chosen EU countries. Polish Journal of Management Studies,
12(1), 37-47.
Coval, J. D., & Moskowitz, T. J. (2001). The geography of investment: informed trading and
asset prices. Journal of Political Economy, 109(4), 811-841.
Dar, A. B., & Maitra, D. (2017). Is gold a weak or strong hedge and safe haven against
stocks? Robust evidences from three major gold-consuming countries. Applied
Economics, 49(53), 5491-5503. https://doi.org/10.1080/00036846.2017.1310998.
Davidson, S., Faff, R., & Hillier, D. (2003). Gold factor exposures in international asset
pricing. Journal of International Financial Markets. Institutions and Money, 13, 271-
289. http://dx.doi.org/10.1016/S1042-4431(02)00048-3.
Diba, B., & Grossman, H. (1984). Rational bubbles in the price of gold. NBER Working
Paper No. 1300. Cambridge, MA: National Bureau of Economic Research.
Dooley, M. P., Isard, P., & Taylor, M. P. (1995). Exchange rates, country-specific shocks and
gold. Applied Financial Economics, 5, 121-129.
Ghosh, D., Levin, E. J., MacMillan, P., & Wright, R. E. (2004). Gold as an inflation hedge?
Studies in Economics and Finance, 22(1), 1-25. http://dx.doi.org/10.1108/eb043380
Ghosh, D., Levin, E. J., Macmillan, P., & Wright, R. E. (2000). Gold as an inflation hedge?
Retrieved from https://www.st-andrews.ac.uk/~wwwecon/papers/dp0021x.pdf
Giannarakis, G., Partalidou, X, Zafeiriou, & Sariannidis, N. (2016). An analysis of United
States on Dow Jones Sustainability Index. Investment Management and Financial
Innovations, 13(3-2), 353-361. doi:10.21511/imfi.13(3-2).2016.07
Eryiğit, M. (2017). Short-term and long-term relationships between gold prices and precious
metal (palladium, silver and platinum) and energy (crude oil and gasoline) prices,
Economic Research-Ekonomska Istraživanja, 30(1), 499-510, DOI:
10.1080/1331677X.2017.1305778.
Ibrahim, M. H. (2012). Financial market risk and gold investment in an emerging market: the
case of Malaysia. International Journal of Islamic and Middle Eastern Finance and
Management, 5(1), 25-34. https://doi.org/10.1108/17538391211216802.
Jaffe, J. (1989). Gold and gold stocks as investments for institutional portfolios. Financial
Analysts Journal, 45, 53-59.
Jurevičienė, D. & Jakanovytė, A. (2015). Alternative investments: valuation of wine as a
means for portfolio diversification. Business: Theory and Management, 16(1), 84-93.
https://doi.org/10.3846/btp.2015.606.
Kolluri, B. R. (1981). Gold as a hedge against inflation: an empirical investigation. Quarterly
Review of Economics and Business, 21, 13-24.
Koutsoyiannis, A. (1983). A short-run pricing model for a speculative asset, tested with data
from the gold bullion market. Applied Economics, 15, 563-581.
Laurent, R. D. (1994). Is there a role for gold in monetary policy? Economic Perspectives, 18,
2-14.
Liu, S. M., & Chou, C. H. (2003). Parities and spread trading in gold and silver markets: a
fractional cointegration analysis. Applied Financial Economics, 13, 879-891.
Lucey, B. M., & Tully, E. (2006a). Seasonality, risk and return in daily COMEX gold and
silver data 1982-2002. Applied Financial Economics, 16, 319-333.
Lucey, B. M., & Tully, E. (2006b). The evolving relationship between gold and silver 1978-
2002: evidence from a dynamic cointegration analysis a note. Applied Financial
Economics Letters, 2, 47-53.
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
263
Mačí, J., & Valentová Hovorková, V. (2017). Loan versus Bond Financing of Czech
Companies and the influence of the Global Recession. Journal of Competitiveness, 9(1), 72-
88. DOI: 10.7441/joc.2017.01.05.
Mahdavi, S., & Zhou, S. (1997). Gold and commodity prices as leading indicators of
inflation: tests of long-run relationship and predictive performance. Journal of
Economics and Business, 49, 475-489.
Makiel, K. (2015). Portfolio diversification during monetary loosening policy. The Journal
of Risk finance, 16(2), 197-214.
Markowitz, H. (1952). Portfolio selection. Journal of Finance, 7(1), 77-91.
Markowitz, H. (1959). Portfolio selection: efficient diversification of investments. New York:
John Wiley & Sons.
Menon, N. (2015). Going for gold. Gurgaon, India: Penguin Publisher.
Mentel, G., Brożyna J. (2015). Compatibility of market risk measures, Journal of
International Studies, 8(2), 52-62. DOI: 10.14254/2071-8330.2015/8-2/5.
Michaud, R. O., Michaud, R., & Pulvermacher, K. (2006). Gold as a strategic asset. London:
World Gold Council.
Moore, G. H. (1990). Gold prices and a leading index of inflation. Challenge, 33, 52-56.
Mouna, A., & Jarboui, A. (2015). Financial literacy and portfolio diversification: an
observation from the Tunisian stock market. International Journal of Bank Marketing,
33(6), 808-822, https://doi.org/10.1108/IJBM-03-2015-0032
Mouselli, S., Aljazaerli, M. A., & Sirop, R. (2016). Corruption and stock market
development: new evidence from GCC countries. Business: Theory and Practice, 17(2),
117-127. https://doi.org/10.3846/btp.2016.555.
Munro, B., & Silberman, K. (2008). Optimal asset allocation in different economic
environments. Cadiz Securities Quantitative Research Report.
Nawrocki, T. (2018). Opportunities and threats associated with an investment in shares of
innovative companies evidence from Polish capital market. Oeconomia
Copernicana, 9(2), 225-244. https://doi.org/10.24136/oc.2018.012.
Nuhiu, A., Hoti, A., & Bektashi, M. (2017). Determinants of commercial banks profitability
through analysis of financial performance indicators: evidence from Kosovo. Business:
Theory and Practice, 18(1), 160-170. https://doi.org/10.3846/btp.2017.017
Ohanyan G., Androniceanu, A. (2017), Evaluation of IMF program effects on employment in
the EU, Acta Oeconomica, 67(3), 311-332.
Pindyck, R. S. (1993). The present value model of rational commodity pricing. The Economic
Journal, 103, 511-530.
Riley, C. (2010). A new gold rush: investing in precious metals. The Journal of Investing,
19(2), 95-100. http://dx.doi.org/10.3905/joi.2010.19.2.095
Shahbaz, M., Iqbal Tahir, M., Ali, I., & Rehman, I. U. (2014). Is gold investment a hedge
against inflation in Pakistan? A co-integration and causality analysis in the presence of
structural breaks. The North American Journal of Economics and Finance, 28, 190-205.
Sherman, E. (1982). Gold: a conservative, prudent diversifier. Journal of Portfolio
Management, 8(3), 21-27. http://dx.doi.org/10.3905/jpm.1982.408850
Sherman, E. (1983). A gold pricing model. Journal of Portfolio Management, 9, 68-70.
Sherman, E. (1986). Gold investment: theory and application. New York: Prentice Hall.
Sjaastad, L. A., & Scacciallani, F. (1996). The price of gold and the exchange rate. Journal of
Money and Finance, 15, 879-897.
Smith, G. (2002). Tests of the random walk hypothesis for London gold prices. Applied
Economics Letters, 9, 671-674.
Ligita Gaspareniene, et. al.
ISSN 2071-789X
INTERDISCIPLINARY APPROACH TO ECONOMICS AND SOCIOLOGY
Economics & Sociology, Vol. 11, No. 3, 2018
264
Śliwiński, P., & Łobza, M. (2017). The impact of global risk on the performance of socially
responsible and conventional stock indices. Equilibrium. Quarterly Journal of
Economics and Economic Policy, 12(4), 657-674. doi: 10.24136/eq.v12i4.34.
The Word Gold Council. (2017). Gold demand trends: full year 2016. Retrieved from
https://www.gold.org/download/file/5463/gdt-fy-2016.pdf
The World Gold Council. (2013). Gold investor. Risk management and capital preservation.
Retrieved from http://www.gold.org/sites/default/files/documents/gold-investor-
201301.pdf
The World Gold Council. (2016). Gold demand trends: full year 2015. Retrieved from
https://www.gold.org/download/file/4815/GDT_Q4_2015.pdf
Vukovic, D. B., & Prosin, V. (2018). The prospective low risk hedge fund capital allocation
line model: evidence from the debt market. Oeconomia Copernicana, 9(3).
http://dx.doi.org/10.24136/oc.2018.018.
Vukovic, D., Hanic, E., & Hanic, H. (2017). Financial integration in the European Union - the
impact of the crisis on the bond market. Equilibrium. Quarterly Journal of Economics
and Economic Policy, 12(2), 195-210. https://doi.org/10.24136/eq.v12i2.10.
Vychytilova, J. (2018). Stock Market Development Beyond THE GFC: The Case of V4
Countries. Journal of Competitiveness, 10(1), 149-163. DOI: 10.7441/joc.2018.02.10
Wang, G. J., Xie, C., Jiang, Z. Q., & Stanley, H. E. (2016). Extreme risk spillover effects in
world gold markets and the global financial crisis. International Review of Economics
and Finance, 46, 55-77. https://doi.org/10.1016/j.iref.2016.08.004
Worthington, A. C. & Pahlavani, M. (2007). Gold investment as an inflationary hedge:
cointegration evidence with allowance for endogenous structural breaks. Applied
Financial Economics Letters, 3(4), 259-262.
https://doi.org/10.1080/17446540601118301
Yu, H.-C., Lee, C.-J. & Shih, T.-L. (2016). Weekday effects on gold: Tokyo, London, and
New York markets. Banks and Bank Systems, 11(2), 33-44.
doi:10.21511/bbs.11(2).2016.04
... Also, crude oil spot prices were found to be intercept stationary at levels using the DF-GLS test of Elliott et al. (1992) by Balcilar et al. (2015) using a 5-day daily spot prices data from January 1987 to February 2012. Similarly, the stochastic properties of gold, by far the most established portfolio risk management tool (Gaspareniene et al., 2018), has been explored by several studies (Gaspareniene et al., 2018;Hassani et al., 2015;Ismail et al., 2009;Rathnayaka and Seneviratna, 2019;Tripathy, 2017). Using monthly data from January 1972 to December 2013, ADF and Bai and Perron (2003) test of structural breaks, Hassani et al. (2015) concluded that, gold prices were non-stationary with structural breaks on 1980(4), 1984(4), 2001(8), 2007(9), and 2007(12). ...
... Also, crude oil spot prices were found to be intercept stationary at levels using the DF-GLS test of Elliott et al. (1992) by Balcilar et al. (2015) using a 5-day daily spot prices data from January 1987 to February 2012. Similarly, the stochastic properties of gold, by far the most established portfolio risk management tool (Gaspareniene et al., 2018), has been explored by several studies (Gaspareniene et al., 2018;Hassani et al., 2015;Ismail et al., 2009;Rathnayaka and Seneviratna, 2019;Tripathy, 2017). Using monthly data from January 1972 to December 2013, ADF and Bai and Perron (2003) test of structural breaks, Hassani et al. (2015) concluded that, gold prices were non-stationary with structural breaks on 1980(4), 1984(4), 2001(8), 2007(9), and 2007(12). ...
Article
The analysis of historical price data for patterns and using such patterns for predictions and policy recommendations has become ubiquitous in the existing economics literature. These predictions and recommendations are premised on the stability of the statistical properties and inter-variable dynamics for which a single regime or few number of regimes can capture. This, however, is a strong assumption with serious repercussions if violated. In this study, the appropriateness of the stability assumption is questioned using various recursive regressions to test stability, consistency of stationarity and stability in inter-variable dynamics between crude oil, gold, silver, and platinum prices. Using monthly data sourced from the World Bank Commodity Price Data (Pink Sheet) from January 1, 9960 to March 2022, our empirical analysis found level prices of oil, gold, and platinum to be consistently non-stationary with rare exceptions. The level price of silver however is found to be inconsistent with multiple regime switches while the logged series of all variables yielded non-stationarity. The default is stationarity for all the variables when price series are logged differenced and/or differenced for oil, silver, and platinum. Differenced gold prices resulted in inconsistent stationarity with multiple regime changes. Even if rare, the stationarity of all the variables is dependent on time and sample size due to the inconsistence in the stationarity verdict. On the bi-variate relationship in the long run, only level silver prices are found to be cointegrated with oil while logged silver prices are inconsistently cointegrated with logged oil prices. Also, in the short-run, only log of oil prices is found to Granger cause log of silver prices. It is thus recommended that researchers and policy makers be tempered in extrapolating statistical findings in general and the price and inter-price dynamics of oil, gold, silver and platinum into the future.
... R Square is the coefficient of determination of a value between 0 and 1, an indicator of good fit. Standard Error is another goodness-of-fit measure that depicts the precision of the regression analysis (Gasparėnienė et al. 2018). ...
Thesis
Gold price research has attracted considerable attention in recent years due to the sharp increasing price trend. The price of gold hit an all-time high in August 2020 as the price increased by more than 30 per cent since January despite the COVID pandemic. The main objective of this research is to examine the determinants of gold price by investigating the relationship of the four key influencing variables affecting the price of gold; the US dollar, inflation, crude oil and silver. Prior studies have primarily focused on one variable. This study contributes to research and practice by producing a comprehensive analysis of various factors and their relationship with gold to gain an in-depth understanding of the driving forces that influence the price of gold. To advance the current literature, quantitative methods were used. Data from 1979 to 2019 was used to calculate the mean, median, maximum, minimum and range for both the dependent and independent variables. A linear regression analysis was then conducted to investigate the correlation between the independent variables and gold price. Keywords: Gold price; Oil Price; Linear Regression; Inflation.
... Gold and Bitcoin can be considered as safe havens during the global pandemic when investing in the NYSE energy stocks. This result is consistent with that of Gaspareniene et al. (2018) and Robiyanto et al. (2019). However, it is inconsistent with that of and Corbet et al. (2020). ...
Article
Full-text available
In this paper, we examine the relationship between the volatilities of the energy index, crude oil, gas prices, and financial assets (Gold, Bitcoin, and G7 stock indexes), especially during the coronavirus crisis. The study tests the presence of regime changes in the GARCH volatility dynamics of the G7 stock indexes, Bitcoin, Gold, and energy assets (energy index, oil, and gas) by using the Markov–Switching GARCH model. It estimates the dynamic correlation and volatility spillover between energy and financial assets, by using the multivariate MSGARCH models. The estimation results of the Markov-Switching-BEKK-GARCH prove the volatility spillover from energy assets to financial assets. For the high regime, the results indicate a high level of dynamic correlation between energy assets and stock indexes which proves the contagion effect of the COVID-19. On the contrary, the dynamic conditional correlation between energy assets and Gold prices decreased during the COVID-19 crisis. This paper makes an original contribution in identifying the contagion between energy and financial assets and indicates that Gold is a safe haven for all energy and financial assets during the COVID-19 crisis. However, Bitcoin cannot be considered as a safe haven during the COVID-19 pandemic when investing in energy assets (crude oil and gas).
... It is an assumption of the method that the data set applied in this method, which provides successful results in short and medium term modelling and prediction studies, is a discrete and stationary data set consisting of observation values with equal time intervals. The basic principle of the Box-Jenkins methodology is based on its value in any period of time series, a combination of past observations and error terms (Gaspareniene, 2018). Since Box-Jenkins models can only be used in stationary series, stationary condition is important in determining the appropriate model group (Anvari et al., 2016). ...
Article
Full-text available
Tourism demand is the basis on which all commercial decisions concerning tourism ultimately depend. Accurate estimation of tourism demand is essential for the tourism industry because it can help reduce risk and uncertainty as well as effectively provide basic information for better tourism planning. The purpose of this study is to develop the optimal forecasting model that yields the highest accuracy when compared to the forecast performances of three different methods, namely Artificial Neural Network (ANN), Exponential Smoothing, and Box-Jenkins methods for forecasting monthly inbound tourist flows to Croatia. Prior studies have been applied to forecast tourism demand to Croatia based on time series models and casual methods. However, the monthly and comparative tourism demand forecasting studies using ANNs are still limited, and this paper aims to fill this gap. The number of monthly foreign tourist arrivals to Croatia covers the period between January 2005-December 2019 data were used to build optimal forecasting models. Forecasting performances of the models were measured by Mean Absolute Percentage Error (MAPE) statistics. As a result of the experiments carried out, when compared to the forecasting performances of various models, 12 lagged ANN models, which have [4-3-1] architecture, were seen to perform best among all models applied in this study. Considering both the empirical findings obtained from this study and previous studies on tourism forecasting, it can be seen that ANN models that do not have any negativities (such as over-training, faulty architecture, etc.) produce successful forecasting results when compared with results generated by conventional statistical methods.
Article
Full-text available
The time series approach has been considered into account for predicting Gold Price in India. The secondary data is taken from the indiastat website from the year 1986 to the year 2018 for the period of 33 years. Error measures such as Root Mean Square Error, Mean Absolute Percentage Error, and Mean Absolute Error are used to validate the model. SARIMA model (0,1,0) (1,0,1) 12 finalized as the best fit for the taken dataset. The study helps the decision makers, investors and commodity traders to take the prior proclamation.
Article
Full-text available
The results of the prediction of business and economic characteristics provide valuable information to stakeholders (business owners and managers, investors, and shareholders). The aim of the paper is to provide a comprehensive overview on methods applied in practice for predicting significant business and economic variables. The research is structured into selected key industries for most of the world's economies. An extensive literary review of the scientific papers published over the last decade revealed that the most used prediction methods include ANN, GARCH combined with ARIMA. These are the methods strong enough to capture the specifics of the industries for the economic and business prediction purposes. The LS-SVM and ARIMA methods are used separately to a lesser extent. The other methods were used mainly for the purpose of vali-dation of their predicting applicability.
Conference Paper
Full-text available
The closure of schools during Covid-19 plague had a pessimistic impact on the performance of the learners preparing for external examinations. A research-designed questionnaire tagged Covid-19 Pandemic Effects on External Examinations of science students’ questionnaire and a Pro-forma were used in gathering the required data for the study. The reliability of the instrument was tested using Cronbach Alpha and a coefficient of 0.856 at .05 alpha level of significance was derived. The study showed there is no significant relationship between the effect of pandemic on external examination and there is a significant relationship between the performance of science students in 2020 WAEC and NECO. It was also revealed that disruption of school academic calendar affected the performance of students during external examination and as well as lack of parental guidance during preparation for external examinations affected the students. Since closure of schools has negative impact on students, government should subsidize the cost of internet services in order for the children of low income earner to have access in case of future sudden school closure. School management, parents, teachers and students should continue to play their roles to sustain the level of students’ academic performance and possible improve upon it, were recommended and concluded among others.
Article
Introduction. Gold price has being demonstrating stable tendency to rise during recent time. Stronger geopolitical tensions support the view that that international political economy factors may play a role driving gold price. In the same time post-crisis global economic uncer­tainty and global expansion of liquidity may affect gold price by itself. The purpose of the article is to find which gold price factor is the most important taking into account assumption that global macrofinancial conditions affect assets prices yet the real economy. Results. We consider competitive approaches on gold price factors: international politi­cal economy, safe haven effect, hedging against monetary shocks, assets with negative beta. Gold reserves accumulation may support ideas that reflected in economic literature about gold price factors. In the same time, fraction of gold in global exchange reserves didn’t change sub­stantially during recent time. For empirical test the four groups of factors were chosen. Each of them are proxy for geopolitical tensions, global economic policy uncertainty, global exchange reserves accumulation, monetary / financial conditions in US. It is found that factors of inter­national political economy are not valid. The most important factors are global economic policy uncertainty and expansion of global liquidity in the form of low long-term US interest rates and global exchange reserves accumulation. Conclusions. Gold price drivers are on the global macrofinancial conditions side. The role of international political economy factors is overvalued.
Article
Full-text available
Research background: The labour market situation is considered to be the most widely discussed part of economic development. However, it should be noted that the unemployment situation of young people (aged 15–24 years) in Poland in general terms seems to be problematic. Overall, the unemployment rate among young people in Poland is significantly higher than the overall unemployment rate in the EU. Moreover, the situation varies greatly across the regions. Purpose of the article: Using multivariate techniques as a theoretical framework, the main goal of the paper is to identify groups of Polish regions that share similar patterns regarding unemployment among young people. To reach this goal, first a set of labour market indicators were selected. Next, the authors compared the labour market situation of young people between the Polish regions in 2005 and in 2014. Finally, the conclusions regarding the conducted analysis are explored. Methods: The initial calculation is based on the concept of the taxonomic measure developed by Hellwig. The final method used to create clusters of objects (across 16 voivodeships of Poland) is cluster analysis. A segmentation of the voivodeships is observed for the years 2005 and 2014, based on selected indicators to determine the labour market situation. The data was gathered from the databases of the Central Statistical Office of Poland and Eurostat. Findings & Value added: Through the exploration of the advantages of multivariate methods, the nature of youth unemployment is revealed in more detail. Indeed, dendrogram analysis divided the voivodeships into five groups, which are characterized by similar features associated with the labour market. It was found that the groups which emerged in 2005 have a different composition of regions than in 2014; this difference seems to be connected with the economic crisis.
Article
Full-text available
Research background: Institutional investors such as: commercial banks, pension funds, and insurance companies are constantly looking for low-risk stable investment opportunities, whereas one of the solutions can be a simulated portfolio. This research takes a look at the incentive to invest in government debt portfolios, as it can outperform the returns of deposit accounts. Purpose of the article: This study considers several classic methods of portfolio constriction and includes the basis of debt instruments that have not been a research topic for a long period of time. At the same time, this paper analyzes the classic methods of modern portfolio theory with a Sharpe ratio as an indicator of efficiency. Methods: The constructed portfolio consists of four elements from different countries: two government obligations and two bond indexes, aiming to employ international diversification. All the data was collected for the period of 12 years in order to represent the consequences of accrued recessions. Findings & Value added: The past two severe financial crises created a higher demand for stable investments, and more investors are ready to compromise a higher return for it. There-fore, the results of this paper represent a simulation of low-risk hedge fund portfolio construction with the use of highly rated debt instruments.
Article
Full-text available
The stock exchanges are widely considered to be pivotal institutions providing the capital for companies and helping them to gain competitive advantage through economies of scale. This paper examines and sheds additional insight into the stock market behaviour of countries of Visegrad Group, namely the Czech Republic, Slovak Republic, Poland and Hungary, beyond the global financial crisis. The study uses data about official stock market indices of the four official stock exchanges in the Visegrad region, between the years 2010-2016. It applies the correlation analysis in attempt to find out whether the indices, namely the BUX Index, PX Index, SAX Index and WIG 20 Index, have similarities in their behaviour, beyond the crisis. The results indicate the V4 stock markets of geographically connected economics exhibit significant similarities in their behaviour, in the post-crisis period. The paper, inter alia, stresses the importance of the stock markets and provides a coherent overview on official V4 stock markets and their official stock market indices.
Article
Full-text available
Research background: Since the Internet bubble, which took place at the turn of XX and XXI century, on the global capital markets, including Poland, one may note a growing interest in companies focusing on innovations and innovativeness. The main driver of this interest is the belief that in a longer term innovations and expenditures on research and development will translate into an increase in competitive advantage, financial results, and subsequently also the market value of companies. On the other hand, the attention should also be paid to the fact that innovative activity has also another, darker, side, which is identified with the far-reaching uncertainty about its final effects and the possibility of incurring losses, especially in financial dimension. At the same time, it should be noted that implementation of investment strategy regarding the shares of innovative companies is quite troublesome because of the lack of unified methodology for assessing corporate innovativeness and large information diversity in this area. Purpose of the article: The investment efficiency analysis of investment strategy regarding shares of companies perceived to be innovative with simultaneous focusing on the different cases of situation development in time. Methods: The research was carried out for companies listed on the main market of the Warsaw Stock Exchange, taking into consideration various time ranges of investment. The efficiency analysis of this investment strategy was conducted in the risk-return outlay with the use of such measures as: accumulated rate of return, arithmetic average rate of return, standard and semi-standard deviation, as well as coefficients of variation and semi-variation of rate of return and their inverses. Findings & Value added: The obtained results show that in shorter periods of time, inves-tors buy expectations connected with innovative companies and therefore, the efficiency of investment in their shares is relatively high, but in the longer term expectations are revised by companies’ financial results, which in turn often negatively affects the investment efficiency.
Article
Full-text available
Research background: In the last decades social responsible investment has evolved into an important and influential investment class. What supports then the development of SRI? The neoclassical approach suggests that the attractiveness of investment should result from the risk-return relationship that is satisfying for the investor. However, the performance analysis of SRI vs. conventional investment, conducted in numerous research papers, often delivers contradictory conclusions. If financial factors could not explain the phenomenon of SRI, nonfinancial factors may have played a decisive role in the formation of modern SRI market. Purpose of the article: The purpose of this paper is to analyze financial investment perfor-mance of socially responsible vs. respective conventional indices in the periods of high, low and unidentified global risk. Therefore, a following research hypothesis was verified: SR indices perform financially better in high-risk periods than in low-risk periods. This hypoth-esis is justified by the assumption that, when selecting SRI, investors go by a longer invest-ment horizon than they do when selecting other investments, not subject to such verification. Methods: Among SR indices, we chose three to compare them with their conventional counterparts: DJSI US vs. DJITR (USA), DJSI Korea vs. KOSPI (South Korea) and Respect Index vs. WIG20TR (Poland). The VIX index was used as the global measure of risk aver-sion. To measure the relative performance of SR and conventional indices in different risk periods, we applied risk-adjusted performance measures, including RSD, Sharpe and Treynor ratios, traditional and asymmetrical CAPM. Findings & Value added: The research shows that conventional and socially responsible indices do not differ statistically in terms of risk and return irrespective of global risk. Our research confirms that the rising, socially responsible, investment market cannot be analyzed only through the prism of simplified rational choices. Additionally, it should be analyzed in terms of moral philosophy and behavioral economics, including the psycho-social features of investors.
Article
Full-text available
The purpose of this study is to elaborate whether the determinants of commercial banks’ profitability affect the financial performance of commercial banks in Kosovo. Performance evaluation of commercial banks in Kosovo is done through measurement of financial performance indicators such as Return on Average Equity (ROAE), Return on Average Assets (ROAA) and Net Interest Margin (NIM). The study identifies the main factors that affect the profitability of commercial banks through analysis of financial time series and panel data of the banking sector in Kosovo. The study presents three models of financial performance analysis which highlight the influencing factors. The models are based on regression analysis, and the obtained results emphasize the relationship between the determinant factors of commercial banks profitability expressed through analysis of financial performance indicators. The study concludes that commercial banks profitability in Kosovo is driven mainly by internal determinant factors such as capital adequacy, asset quality and management efficiency, while macroeconomic factors have insignificant impact on financial performance of commercial banks.
Article
Full-text available
Research background: In our paper we have analyzed the influence of the crisis on the financial integration in the European Monetary Union. We have analyzed EMU capital market to show the impact of the crisis, with the focus on the bonds market. The determinants of the research are yields and standard deviations on medium-term and long-term triple-A bond markets, as well as CDS medium-term premiums. Purpose of the article: The aim of this paper is to show the volatility of researched deter-minants in periods of crisis in EMU zones.Methods: As a model we used a modified theoretical CAL portfolio model. In the last fifteen years Europe has been faced with two major crises: the world economic crisis and sovereign debt crisis.Findings & Value added: We believe that the sovereign crisis hit EMU more, leaving the deeper implications on the financial integration. Our analysis has showed that the crisis had a major impact on the financial integration. Yields and standard deviations increased multiply in periods of crisis and left the impact of volatility on the capital market. However, the degree of convergence of euro area bond markets largely stabilized in last two years.
Article
The International Monetary Fund (IMF) has undergone notable changes by starting collaboration with the European Union (EU). Hence, this paper seeks to estimate the effects of IMF programmes on employment, with data from the EU-28 between 1993 and 2013. In order to control for selection on observable and unobservable variables, the study employs Propensity Score Matching in combination with the Differences-in-Difference estimator. Next, the robustness of the findings is checked by applying four different matching algorithms. Our paper concludes that employment decreases once a country resorts to the IMF, and this impact is still measurable after two years of programme initiation.
Article
In this article, we test nexus between gold and stocks for the three major gold consumers by using the range of methodologies. First, we assess if there is any time-varying correlation between the two assets. We fail to find any significant time-varying correlation between gold and stock returns in India and the United States. Second, we attempted to investigate the safe-haven property of gold by analysing the decile-wise conditional correlation between stock returns and gold returns at different deciles of stock returns. Third, in order to test the robustness of the results drawn from the decile-wise correlation, we employ wavelet coherence in continuous wavelet framework to test the time and frequency varying nexus between the pair of assets. The range of methodologies employed seems to indicate the weak hedge and safe haven-property of gold for stocks.