The Journal of Portfolio Management

Print ISSN: 0095-4918
In this paper we examine global tactical asset allocation (GTAA) strategies across a broad range of asset classes. Contrary to market timing for single asset classes and tactical allocation across similar assets, this topic has received little attention in the existing literature. Our main finding is that momentum and value strategies applied to GTAA across twelve asset classes deliver statistically and economically significant abnormal returns. For a long top-quartile and short bottom-quartile portfolio based on a combination of momentum and value signals we find a return of 12% per annum over the 1986-2007 period. Performance is stable over time, also present in an out-of-sample period and sufficiently high to overcome transaction costs in practice. The return cannot be explained by potential structural biases towards asset classes with high risk premiums, nor the Fama French and Carhart hedge factors. We argue that financial markets may be macro inefficient due to insufficient ‘smart money’ being available to arbitrage mispricing effects away.
Guo and Savickas [2005] show that aggregate stock market volatility and average idiosyncratic stock volatility jointly forecast stock returns. In this paper, we quantify the economic significance of their results from the perspective of a portfolio manager. That is, we evaluate the performance, e.g., the Sharpe ratio and Jensen's alpha, of a mean-variance manager who tries to time the market based on those two variables. We find that, over the period 1968-2004, the associated market-timing strategy outperforms the buy-and-hold strategy, and the difference is statistically and economically significant.
Performance of investment managers is predominantly evaluated against targeted benchmarks, such as stock, bond or commodity indices. However, most professional databases do not retain timeseries for companies that disappeared, and do not necessarily track the change of constitution in these benchmarks. Consequently, standard tests of performance suffer from the “look-ahead benchmark bias,” where a given strategy is naively back-tested against the assets constituting the benchmark of reference at the end of the testing period (i.e. now), rather than at the very beginning of that period. We report that the “look-ahead benchmark bias” can exhibit a surprisingly large amplitude for portfolios of common stocks (up to 8% per annum for the S&P500 taken as the benchmark), while most studies have emphasized related survival biases in performance of mutual and hedge funds for which the biases can be expected to be even larger. We use the CRSP database from 1926 to 2006 and analyze the running top 500 US capitalizations to demonstrate that this bias can account for a gross overestimation of performance metrics such as the Sharpe ratio as well as an underestimation of risk, as measured for instance by peak-to-valley drawdowns. We demonstrate the presence of a significant bias in the estimation of the survival and look-ahead biases studied in the literature. A general methodology to test the properties of investment strategies is advanced in terms of random strategies with similar investment constraints.
Equilibrium Expected Real Rates of Return for Different Assumptions About Proportions of Assets Outstanding
The recent introduction of CPI-linked bonds by several financial institutions is a milestone in the history of the U.S. financial system. It has potentially far-reaching effects on individual and institutional asset allocation decisions because these securities represent the only true long-run hedge against inflation risk. CPI-linked bonds make possible the creation of additional financial innovations that would use them as the asset base. One such innovation that seems likely is inflation-protected retirement annuities. The introduction of index-linked bonds eliminates one of the main obstacles to the indexation of benefits in private pension plans. A firm could hedge the risk associated with a long-term indexed liability by investing in index-linked bonds with the same duration as the indexed liabilities.
Correlation matrix Monthly Returns, US, February 1990-August 2008
Co-kurtosis matrix Monthly Returns, US, February 1990 -August 2008
This paper examines the advantages of incorporating equity volatility exposure into a global balanced portfolio. We consider two sets of strategies: long implied volatility and long volatility risk premium strategies. To calibrate and assess the risk/return profile of the portfolio, we present an analytical framework, offering pragmatic solutions for long-term investors seeking exposure to volatility. The benefit of volatility exposure for a conventional portfolio is shown through a mean / Value-at-Risk portfolio optimization. Pure volatility investment makes it possible to partially hedge downside equity risk, thus reducing the risk profile of the portfolio. Investing in the volatility risk premium substantially increases returns for a given level of risk. A well calibrated combination of the two strategies allows for enhanced absolute and risk-adjusted returns for the portfolio.
Basu and Drew (in the JPM Spring 2009 issue) argue that lifecycle asset allocation strategies are counterproductive to the retirement savings goals of typical individual investors. Because of the portfolio size effect, most portfolio growth will occur in the years just before retirement when lifecycle funds have already switched to a more conservative asset allocation. In this article, we use the same methodology as Basu and Drew, but we do not share their conclusion that the portfolio size effect soundly overturns the justification for the lifecycle asset allocation strategy. While strategies that maintain a large allocation to stocks do provide many attractive features, we aim to demonstrate that a case for supporting a lifecycle strategy can still be made with modest assumptions for risk aversion and diminishing utility from wealth. Our differing conclusion results from four factors: (1) we compare the interactions between different strategies; (2) we consider a more realistic example for the lifecycle asset allocation strategy; (3) we examine the results for 17 countries; and (4) we provide an expected utility framework to compare different strategies. We find that with a very reasonable degree of risk aversion, investors have reason to prefer the lifecycle strategy in spite of the portfolio size effect.
Risk budgeting interpreted as efficient portfolio allocation is often based on expected outperformance, alpha or information ratio. Once these crucial parameters have been estimated, they are being treated as fixed. In this paper we develop some sense, both theoretical and practical, on the magnitude of the uncertainty and present a method to cape with uncertainty of the expected active returns. We will use examples to demonstrate the impact of the uncertainty. The developed model is widely applicable and can be used to make an optimal risk allocation on different levels of investment strategy (as set allocation, styles, managers, etc.).
We introduce a new dynamic trading strategy based on the systematic misspricing of U.S. companies sponsoring Defined Benefit pension plans. This portfolio produces an average return of 1.51% monthly between 1989 and 2004, with a Sharpe Ratio of 0.26. The returns of the strategy are not explained by those of primary assets. These returns are not related to those of benchmarks in the alternative investments industry either. Hence, we are in the presence of a "pure alpha" strategy that can be ported into a large variety of portfolios to significantly enhance their performance.
Do terrorism-related investing strategies lead to superior investment performance? This study evaluates the risks and returns to two different terrorism-related investment strategies in the U.S. markets over the period from 1994-2006. The first strategy evaluates a sub-portfolio of S&P 500 stocks constructed on the basis of terrorism-related risk scores that measure their operations in countries with a high incidence of terrorism-related activity. The second strategy evaluates a 'terror-free' sub-portfolio of S&P 500 stocks in which stocks are screened if they have operations in countries that the U.S. Department of State has designated as state-sponsors of terrorism. I find that the terrorism-related risk exposure portfolio would have earned, on average, an economically small and statistically insignificant 16 basis point premium per month with a tracking error of 2.8% per month and that of the terror-free portfolio an even smaller -1.6 basis point premium per month with a tracking error of 25 basis points per month. Return attribution analysis using a multi-factor model uncovers interesting differences in systematic exposures to market risks, and factors related to size, market-to-book ratios and momentum.
Because much of the value of equity depends on the option characteristics of investment projects, it is not feasible to calculate equity duration directly. As a result, recent literature has focused on estimating equity duration empirically. By using 25 size and book-to-market portfolios, this paper shows that estimates of equity duration are critically dependent on the specifications of the regression model used to estimate equity duration. In particular, including all three Fama-French factors in the regression can have a dramatic impact on the estimate coefficients.
This paper investigates the risk-return relationship in determination of housing asset pricing. In so doing, the paper evaluates behavioral hypotheses advanced by Case and Shiller (1988, 2002, 2009) in studies of boom and post-boom housing markets. The paper specifies and tests a housing asset pricing model (H-CAPM), whereby expected returns of metropolitan-specific housing markets are equated to the market return, as represented by aggregate US house price time-series. We augment the model by examining the impact of additional risk factors including aggregate stock market returns, idiosyncratic risk, momentum, and Metropolitan Statistical Area (MSA) size effects. Further, we test the robustness of H-CAPM results to inclusion of controls for socioeconomic variables commonly represented in the house price literature, including changes in employment, affordability, and foreclosure incidence. Consistent with the traditional CAPM, we find a sizable and statistically significant influence of the market factor on MSA house price returns. Moreover we show that market betas have varied substantially over time. Also, we find the basic housing CAPM results are robust to the inclusion of other explanatory variables, including standard measures of risk and other housing market fundamentals. Additional tests of the validity of the model using the Fama-MacBeth framework offer further strong support of a positive risk and return relationship in housing. Our findings are supportive of the application of a housing investment risk-return framework in explanation of variation in metro-area cross-section and time-series US house price returns. Further, results strongly corroborate Case-Shiller behavioral research indicating the importance of speculative forces in the determination of U.S. housing returns.
In this paper, we estimate the behavioral component of the Grinblatt and Han (2002) model and derive several testable implications about the expected relationship between the preponderance of disposition - prone investors in a market and volume, volatility and stock returns. To do this, we use a large sample of individual accounts over a six-year period in the 1990's in order to identify investors who are subject to the disposition effect. We then use their trading behavior to construct behavioral factors. We show that when the fraction of "irrational" investor purchases in a stock increases, the unexplained portion of the market price of the stock decreases. We further show that statistical exposure to a disposition factor explains cross-sectional differences in daily returns, controlling for a host of other factors and characteristics. The evidence is consistent with the hypothesis that trade between disposition-prone investors and their counter-parties impact relative prices.
There are no established benchmarks for evaluating currency investment manager performance. Some analysts have suggested that known investing styles like momentum, purchasing power parity, and carry serve as benchmarks. Challenges for this approach include: there is no market portfolio; there are many alternative generic factor constructions; different constructions of the same factor may have low correlations; the 3 factors may not provide diversification; and there is no “buy and hold” in the FX market. An evaluation of professional currency managers’ returns indicates that they are often generated independently from the generic style factors. Skill in timing is what investors should pay for and some managers demonstrate superior skill in timing the factors. Managers are also skilled at minimizing drawdowns relative to the generic factors. The use of generic style factors may be a worst case scenario instead of returns to which an FX investor may aspire.
For decades the Capital Asset Pricing Model (CAPM) has been held as an article of faith among financial economists. The model, usually attributed to 1990 Nobel Laureate William Sharpe (1964), was also developed by Fischer Black (1972), John Lintner (1965), Jan Mossin (1966), and Jack Treyor (1965). CAPM attempted to quantify the relationship between risk and return. Both economists and financial practitioners have long believed that riskier assets must yield a higher expected rate of return to induce investors to hold them. The innovation of CAPM was to specify the particular risk measure that would be priced in the market.
Recent academic work has developed a method to determine, in real time, if a given stock is exhibiting a price bubble. Currently there is speculation in the financial press concerning the existence of a price bubble in the aftermath of the recent IPO of LinkedIn. We analyze stock price tick data from the short lifetime of this stock through May 24, 2011, and we find that LinkedIn has a price bubble.
Despite its power as a transactions network, scant attention has been given to incorporating an electronic call into a major market center such as the NYSE or Nasdaq. An electronic call clears the markets for all assets at predetermined points in time. By bunching many transactions together, a call market increases liquidity, thereby decreasing transaction costs for public participants. After describing alternative call market structures and their attributes, we propose that an open book electronic call be held three times during the trading day: at the open, at 12:00 noon, and at the close. We discuss the impact of this innovation on an array of issues, including order flow and handling, information revelation, and market transparency. We also discuss the proposed changes from the perspectives of investors, listed companies, exchanges, brokers, and regulators.
"Best execution," an obligation on equity market professionals that has been a holy grail in the United States since enactment of the Securities Acts Amendments of 1975, is a multifaceted concept that is difficult to define - and even more challenging to measure. The authors delineate the various measurement and implementation problems. They consider advice for the buy-side trader, implications for the providers of trading services, and caveats with regard to public policy. The best execution obligation in fact is shared across buy-side institutions, broker/dealer intermediaries, and the market centers themselves.
Assessing the price evolution of houses on the basis of average sales prices, as is current practice in Belgium, might be misleading due to changing characteristics of the houses sold in the periods observed. A hedonic index which takes into account changes in characteristics is more appropriate. We use the budget surveys of the Belgian Statistical Institute to illustrate how this also applies for Belgium. The estimated hedonic price index for house sales on the secondary market is practically always below the index based on average sales values for the period considered. This demonstrates the need to collect more extensive data on the characteristics of the dwellings sold in Belgium.
Real Construction Costs, 80,000ft 2 Office Building, R.S. Means Company Data 
Figure A.3: Net Rental Growth on NCREIF Properties 
Recent sharp declines in owner-occupied housing prices naturally raise the question of whether something similar will happen to income-producing properties. It already has based on the nearly 60% decline in the share prices of publicly-traded, commercial property firms from their peak in early 2007. The core model of spatial equilibrium in urban economics suggests this should not be a surprise, as it shows that both real estate sectors are driven by common fundamentals, which should make them perform similarly. On the other hand, stronger limits to arbitrage in housing suggest wider swings in prices unrelated to fundamentals are feasible in that property sector. The data find many more similarities than differences across the two real estate sectors. The simple correlation between appreciation rates on owner-occupied housing and commercial real estate is nearly 40%. Both sectors also exhibit similar time series patterns in their price appreciation, with there being persistence across individual years and mean reversion over longer periods. Commercial real estate capital structure looks to be quite weak due to high leverage combined with strong mean reversion in prices. The aggregate loan-to-value ratio on income-producing properties is about 75%. Estimated mean reversion in price appreciation of at least 25% over relatively short horizons suggests that normal change from the recent peak will leave little or no equity on average to cushion against any future negative shocks.
This paper develops a new tool for discovering mispriced securities based on an analysis of comovement in asses prices. Recent research in finance has demonstrated that comovement can be due to the trading patterns of noise traders as well as underlying economic fundamentals. Because comovement can be measured much more accurately than expected returns, it can be used to identify securities for which the influence of noise traders is high. Those are situations in which mispricing is most likely to exist. Therefore, analysis of comovement can provide important information about potential mispricing.
The objective of this paper is to identify the determinants of office capitalization rates for a panel of 52 countries (developed and emerging countries) between 2000 and 2006. Our assumption, based on Capital Asset Pricing Model, is that the capitalization rate should be at least proportional to the country’s risk perception, as measured by the risk premium on the 10-year government bond yield. Because of the endogeneity of the latter variable, our empirical methodology requires that we estimate first a model explaining the 10-year bond yield. It will be the occasion to discuss the determinants of the risk premium on the bond market. Using a SURE random effect Hausman-Taylor estimator (Hausman & Taylor, 1981), w also take into account the possible correlation between the country risk characteristics on the bond markets and those that determine the real estate market. Our results show that government bond yield is the main determinant of the capitalization rate. We estimate that 1 percentage point increase in the government bond yield will raisse the capitalization rate by about 0.19 percentage point. Real estate variables play also a role, but to a lesser extent. Turning to determinants of the 10-year bond yield, macroeconomic fundamentals are significant determinants of the country risk premium, especially the capacity to honor short-term financial engagements. In addition, the country’s risk history has also very important effect on the investors’ current risk perception.
The central message of this paper is that nobody should be using the sample covariance matrix for the purpose of portfolio optimization. It contains estimation error of the kind most likely to perturb a mean-variance optimizer. In its place, we suggest using the matrix obtained from the sample covariance matrix through a transformation called shrinkage. This tends to pull the most extreme coefficients towards more central values, thereby systematically reducing estimation error where it matters most. Statistically, the challenge is to know the optimal shrinkage intensity, and we give the formula for that. Without changing any other step in the portfolio optimization process, we show on actual stock market data that shrinkage reduces tracking error relative to a benchmark index, and substantially increases the realized information ratio of the active portfolio manager.
We investigate the efficacy of riding the yield curve. This strategy dictates holding longer-term treasury bills when the yield curve is upwardsloping. We find that the strategy is surprisingly effective. it stochastically dominates buying and holding shorter-term bills for large subperiods, and nearly dominates for the entire sample period, 1949-1988. Our empirical results suggest that abnormal profit opportunities are available from selectively increasing the maturity of a short-term portfolio.
Most institutional and individual portfolios are very undiversified in real estate: many hold no real estate at all, many have holdings highly concentrated in certain regions or types of real estate. The risk of these concentrated holdings is not hedged. We propose here that cash-settled futures and options markets be opened on real estate to better allow diversification and hedging, and show that these markets solve problems that have hampered other real estate hedging media in the past. Related institutions, such as home equity insurance, might develop around the futures and options markets. The establishment of these markets is likely to increase the quantity of reproducible real estate, and lower rents on real estate. It may also reduce the amplitude of speculative real estate price movements and dampen the business cycle.
Transaction costs in trading involve both risk and return. The return is associated with the cost of immediate execution and the risk is a result of price movements during a more gradual trading. The paper shows that the trade-off between risk and return in optimal execution should reflect the same risk preferences as in ordinary investment. The paper develops models of the joint optimization of positions and trades, and shows conditions under which optimal execution does not depend upon the other holdings in the portfolio. Optimal execution however may involve trades in assets other than those listed in the order; these can hedge the trading risks. The implications of the model for trading with reversals and continuations are developed. The model implies a natural measure of liquidity risk
Who should buy options (ordinary or "exotic"), and who should sell? Buyers and sellers must differ from the average investor, who will not undertake options positions. We develop a simple binomial model to characterize the expectations (relative to the average or consensus) which must be held by investors to justify buying or selling various types of derivatives, or following dynamic strategies which generate similar payoffs. European option sellers must believe markets are more mean-reverting than average; option buyers must believe they are more mean-averting. The probabilities of ordinary option buyers and sellers are path independent and their expected return process must be a martingale. Path-dependent options or dynamic strategies imply probabilities which are path dependent. "Asian" derivative purchasers must believe the expected return to the underlying asset decreases through time. Lookback purchasers believe the opposite.
Business cycles in different regions of the United States tend to synchronize. This study investigates the reasons behind this synchronization of business cycles and the consequent formation of a national business cycle. Trade between regions may not be strong enough for one region to "drive" business cycle fluctuations in another region. This study suggests that regional business cycles synchronize due to a nonlinear "mode-locking" process in which weakly coupled oscillating systems (regions) tend to synchronize. There is no definitive test for mode-lock. However, simulations, correlations, Granger causality tests, tests for nonlinearities, vector autoregressions, and spectral analysis reveal modest econometric support for the regional mode-locking hypothesis of business cycle synchronization. Copyright Blackwell Publishers, 2005
Is there any justification for investing in managed mutual funds, or are managed funds for suckers, as indexing advocates argue? We answer this question by looking at a long time span of real fund returns (27 years) for one specific company (Vanguard) that is notable for its low fees on managed funds. By creating synthetic portfolios—portfolios based on the assets of Vanguard’s mutual funds—we find that whether index funds or managed funds are the superior buy depends on the time span in question, but that managed funds almost always have a lower standard deviation of return than index funds.
It is well known that the voluntary reporting of hedge funds may cause biases in estimates of their investment returns. But wide disagreements exist in explaining why hedge funds stop reporting to the datagathering services. Academic studies have suggested that poor or failing funds stop reporting while industry analysts suggest that better performing funds cease reporting because they no longer need to attract new capital. Using the TASS dataset, we find that hedge funds’ returns are significantly worse at the end of their reporting live. We then use survival time analysis techniques to examine the funds’ time to failure and changes in the hazard rate (i.e., the probability of failure) over time. We also estimate the effects of funds’ performance, size, and other characteristics on the hazard rate. Consistent with the finding on funds’ returns at the end of their reporting lives, we find that better performing and larger hedge funds have lower hazard rates.
The popular perception is that hedge funds follow a reasonably well defined market-neutral investment style. While this long- short investment strategy may have characterized the first hedge funds, today hedge funds are a reasonably heterogeneous group. They are better defined in terms of their freedom from the constraints imposed by the Investment Company Act of 1940, than they are by the particular style of investment. We study the monthly return history of hedge funds over the period 1989 through to January 2000 and find that there are in fact a number of distinct styles of management. We find that differences in investment style contribute about 20 percent of the cross sectional variability in hedge fund performance. This result is consistent across the years of our sample and is robust to the way in which we determine investment style. We conclude that appropriate style analysis and style management are crucial to success for investors looking to invest in this market.
Changes in variance, or volatility, over time can be modeled using the approach based on autoregressive conditional heteroscedasticity. Another approach is to model variance as an unobserved stochastic process. Although it is not easy to obtain the exact likelihood function for such stochastic variance models, they tie in closely with developments in finance theory and have certain statistical attractions. This article sets up a multivariate model, discusses its statistical treatment, and shows how it can be modified to capture common movements in volatility in a very natural way. The model is then fitted to daily observations on exchange rates.
I modify the uniform-price auction rules in allowing the seller to ration bidders. This allows me to provide a strategic foundation for underpricing when the seller has an interest in ownership dispersion. Moreover, many of the so-called "collusive-seeming" equilibria disappear.
With the population in the U.S. and other countries ageing rapidly, the burden of future pension liabilities is ever increasing. In recent years, governments and companies have become much more aware of the inherent risks that are involved. As a consequence, there is a worldwide tendency to shift from defined benefit pension plans to defined contribution plans. The implications for employees are far-reaching: under a defined contribution plan, the employee bears the investment risk: the level of his pension depends on the return on his investments. Under a defined benefit system, the level of pensions is fixed and the sponsor (in many cases the employer) bears the investment risk: the premiums required to fund the pension depend on the return on investments. In this joumal, Bodie and Crane (1999) (BC) recognize that the transfer of investment risk from employer to employee calls for easy-to-implement investment strategies that correctly reflect the trade-off between the risk of a poor pension and the joy of a sumptuous pension. They compare investments in traditional equity and bonds with investments in TIPS (inflation linked bonds) and equity with a protective floor. Their results suggest that a series of investments in a product with a protective floor have a much higher chance of reaching a specified retirement income level than investments in a mixture of equity and fixed income securities.
The US economy is arguably following an unsustainable trajectory. The main indicators of this are a large current account deficit, a large federal budget deficit and trend-wise increasing costs of Social Security and Medicare. In this chapter, we will discuss these observations and to what extent the financial and economic crisis may have changed the outlook. Before this, we need to define what we mean by sustainability. An often used definition of sustainability is that the inter-temporal budget restriction is satisfied.
In this paper we argue that book-to-market and size attributes represent sensitivities of firm returns to several risk factors, and in so doing they subsume the information in other attributes. Although this gives them high cross-sectional explanatory power, they are not very indicative if we are concerned with testing whether an individual risk factor is priced. In that regard, claiming that financial distress is not priced, by only considering probability of bankruptcy, seems premature. Rational investors may also care about recovery rates and the relatively higher mean returns observed for small firms with very low book-to-market ratios is consistent with this view. To analyse recovery risk, we construct mimicking portfolios by sorting stocks on less noisy attributes such as fixed-assets and intangible-assets ratios. We find that recovery risk mimicking portfolios exhibit typical risk factor characteristics, and perform well in explaining the cross-section of returns. The results suggest that recovery risk factor is a good candidate to be priced, and much of the explanatory power of the size attribute comes from the fact that it embodies useful information regarding recovery risk. Overall, our findings have important portfolio management implications.
We present empirical evidence that stocks with low volatility earn high risk-adjusted returns. The annual alpha spread of global low versus high volatility decile portfolios amounts to 12% over the 1986-2006 period. We also observe this volatility effect within the US, European and Japanese markets in isolation. Furthermore, we find that the volatility effect cannot be explained by other well-known effects such as value and size. Our results indicate that equity investors overpay for risky stocks. Possible explanations for this phenomenon include (i) leverage restrictions, (ii) inefficient two-step investment processes, and (iii) behavioral biases of private investors. In order to exploit the volatility effect in practice we argue that investors should include low risk stocks as a separate asset class in the strategic asset allocation phase of their investment process.
Active portfolio management is commonly partitioned into two types ofactivities: market timing, which requires forecasts of broad-based marketmovements, and security analysis, which requires the selection of individualstocks that are perceived to be underpriced by the market. Merton (1981) hasprovided an inciteful and easily-implemented means to place a value on markettiming skills. In contrast, while a normative theory of stock selection wasoutlined long ago in Treynor and Black's (1973) work, no convenient means ofvaluing potential selection ability has yet been devised.We present a framework in which the value of a security analyst can becomputed. We also treat market timing ability in this framework, andtherefore can compare the relative values of each type of investmentanalysts. We find that stock selection is potentially extremely valuable, butthat its value depends critically on the forecast interval, on the correlationstructure of residual stock returns, and on the ability to engage in shortsales. Finally, we show how to modify the value of selection for theimportant case in which analysts' forecasts of stocks' alphas are subject toerror.
Regression of Annual Wage Growth (in percentage terms) on annual return to U.S. 30 Day Treasury Bills. The estimated model estimates are reported below the X axis. The coefficient estimate is statistically significant at greater than the 1% level. Observations are indicated by boxes. Larger boxes are more proximate in time. Date labels indicate the month and year of the observation. 
Summary Statistics of the Series'
Efficient Frontier 
Correlation Among Series'
Portfolio Weights 
We introduce a new hybrid approach to joint estimation of Value at Risk (VaR) and Expected Shortfall (ES) for high quantiles of return distributions. We investigate the relative performance of VaR and ES models using daily returns for sixteen stock market indices (eight from developed and eight from emerging markets) prior to and during the 2008 financial crisis. In addition to widely used VaR and ES models, we also study the behavior of conditional and unconditional extreme value (EV) models to generate 99 percent confidence level estimates as well as developing a new loss function that relates tail losses to ES forecasts. Backtesting results show that only our proposed new hybrid and Extreme Value (EV)-based VaR models provide adequate protection in both developed and emerging markets, but that the hybrid approach does this at a significantly lower cost in capital reserves. In ES estimation the hybrid model yields the smallest error statistics surpassing even the EV models, especially in the developed markets.
I identify the mechanism by which the value premium is realized by index investors by examining the per-share earnings of the S&P/Barra 500 Growth and Value indices from 1976 to 2000 and their returns from 1974 to 2002. Over this period, the growth in per-share earnings and the price return for these two indices are virtually identical. It follows that the difference in their total return can be attributed entirely to the difference in their free cash flow yield. I show that this result is not an artifact of data mining, but an inescapable consequence of the regular rebalancing that these indices undergo, and propose a simple positive model of corporate earnings growth that predicts it. There is a rich literature on why the value premium exists, but little clarity on how it is realized. I identify free cash flow (which is composed of dividends, share buybacks and takeovers) as the mechanism by which the value premium is realized by index investors. Finally, I highlight a curious aspect of the risk of these two indices. The per-share earnings of the S&P/Barra 500 Value Index are more volatile than those of the S&P/Barra 500 Growth Index, while the returns of the S&P/Barra 500 Growth Index are more volatile than those of the S&P/Barra 500 Value Index.
We develop a daily measure of average stock variance and study whether it can predict market returns one day ahead. Using a time-invariant prediction model we find a robust predictive relation between these variables which cannot be used to profitably time the market. A closer look reveals that the strength and even the direction of the predictive relation vary significantly over short periods of time. Moreover, a simple timing strategy that exploits this variation over time significantly outperforms the market buy-and-hold strategy in terms of the mean-variance tradeoff. The evidence shows that predictability is stronger during business-cycle contractions and that our timing strategy is profitable because it avoids losses during bad times. Last, parameter breaks occur very frequently over short periods of time, and not only when the economy switches the phase of the business cycle. Our results suggest that idiosyncratic risk matters in asset pricing and that its effect is time varying.
Sobol and Halton runs for tranche A and two Monte Carlo runs using RAN2
Sobol and Halton runs for tranche A and an average of twenty Monte Carlo runs using RAN2
this paper were obtained using FINDER. One of the improvements was developing the table of primitive polynomials and initial direction numbers for dimensions up to 360.
Any robust institutional investment process must be complemented with a proper risk management policy. The vast array of possible trading strategies and ability to employ leverage and derivatives makes evaluating risk associated to hedge fund investments, and for that matter, a portfolio of hedge funds a rather herculean task. In our experiences, an Investment Committee’s due diligence on a hedge fund typically consists of following a pre-determined process. While this process may vary across investors, it frequently includes examining the portfolio manager’s background, the organizational structure of the fund, the legal documents (e.g. Private Placement Memorandum, subscription documents, etc.), and the internal risk management policies. The examination of the risk management policy often includes looking at a current portfolio’s risk characteristics as summarized on recent prime-broker or third party risk aggregation reports, or even reviewing the actual underlying positions. Additionally, policy limits may exist on the position, sector/industry, and gross/net exposures. Finally, the quants have put their fingerprints on the plethora of other risk measurements, such as VaR, conditional VaR, stress tests, dv01s, Greeks, etc. We feel that most of this analysis, while important, misses the critical issues of a proper risk management policy: (1) Actionable Risk Management, (2) Fraud Responsibilities, and (3) Hidden Risks.
Investors who are invested in (or bear responsibility for) many active portfolios face a resource allocation problem: To which products should they direct their attention and scrutiny? Ideally they will focus their attention on portfolios that appear to be in trouble, but these are not easily identified using classical methods of performance evaluation. In fact, it is often claimed that it takes forty years to determine whether an active portfolio outperforms its benchmark. The claim is fallacious. In this article, we show how a statistical process control scheme known as the CUSUM, which is closely related to Wald's [1947] Sequential Probability Ratio Test, can be used to reliably detect flat-to-the-benchmark performance in forty months, and underperformance faster still. By rapidly detecting underperformance, the CUSUM allows investors to focus their attention on potential problems before they have a serious impact on the performance of the overall portfolio. The CUSUM procedure is provably optimal: For any given rate of false alarms, no other procedure can detect underperformance faster. It is robust to the distribution of excess returns, allowing its use in almost any asset class, including equities, fixed income, currencies and hedge funds without modification, and is currently being used to monitor over $500 billion in actively managed assets.
As quantitative techniques have become commonplace in the investment industry, the mitigation of estimation and model risk in portfolio management has grown in importance. Robust optimization, which incorporates estimation error directly into the portfolio optimization process, is typically used with conventional robust statistical estimation methods. This perspective on the robust optimization approach reviews useful practical extensions and discusses potential applications for robust portfolio optimization.
Top-cited authors
Robert E. Whaley
  • Vanderbilt University
William Ziemba
  • University of British Columbia - Vancouver
William N. Goetzmann
  • Yale University
Meir Statman
  • Santa Clara University
Richard Roll
  • California Institute of Technology