Variable Selection for Portfolio Choice

We study asset allocation when the conditional moments of returns are partly predictable. Rather than first model the return distribution and subsequently characterize the portfolio choice, we determine directly the dependence of the optimal portfolio weights on the predictive variables. We combine the predictors into a single index that best captures time variations in investment opportunities. This index helps investors determine which economic variables they should track and, more importantly, in what combination. We consider investors with both expected utility (mean variance and CRRA) and nonexpected utility (ambiguity aversion and prospect theory) objectives and characterize their market timing, horizon effects, and hedging demands.     

Time to wealth goals in capital accumulation

This paper considers the problem of investment of capital in risky assets in a dynamic capital market in continuous time. The model controls risk, and in particular the risk associated with errors in the estimation of asset returns. The framework for investment risk is a geometric Brownian motion model for asset prices, with random rates of return. The information filtration process and the capital allocation decisions are considered separately. The filtration is based on a Bayesian model for asset prices, and an (empirical) Bayes estimator for current price dynamics is developed from the price history. Given the conditional price dynamics, investors allocate wealth to achieve their financial goals efficiently over time. The price updating and wealth reallocations occur when control limits on the wealth process are attained. A Bayesian fractional Kelly strategy is optimal at each rebalancing, assuming that the risky assets are jointly lognormal distributed. The strategy minimizes the expected time to the upper wealth limit while maintaining a high probability of reaching that goal before falling to a lower wealth limit. The fractional Kelly strategy is a blend of the log-optimal portfolio and cash and is equivalently represented by a negative power utility function, under the multivariate lognormal distribution assumption. By rebalancing when control limits are reached, the wealth goals approach provides greater control over downside risk and upside growth. The wealth goals approach with random rebalancing times is compared to the expected utility approach with fixed rebalancing times in an asset allocation problem involving stocks, bonds, and cash.     

Optimal Estimation of the Risk Premium for the Long Run and Asset Allocation: A Case of Compounded Estimation Risk

It is well known that an unbiased forecast of the terminal valueof a portfolio requires compounding at the arithmetic mean returnover the investment horizon. However, the maximum-likelihoodpractice, common with academics, of compounding at the estimatorof mean return results in upward biased and highly inefficientestimates of long-term expected returns. We derive analyticallyboth an unbiased and a small-sample efficient estimator of long-termexpected returns for a given sample size and horizon. Both estimatorsentail penalties that reduce the annual compounding rate asthe investment horizon increases. The unbiased estimator, whichis far lower than the compounded arithmetic average, is stillvery inefficient, often more so than a simple geometric estimatorknown to practitioners. Our small-sample efficient estimatoris even lower. These results compound the sobering evidencein recent work that the equity risk premium is lower than suggestedby post-1926 data. Our methodology and results are robust toextensions such as predictable returns. We also confirm analyticallythat parameter uncertainty, properly incorporated, producesoptimal asset allocations, in stark contrast to conventionalwisdom. Longer investment horizons require lower, not higher,allocations to risky assets.     

Optimal Portfolio Choice for Long-Horizon Investors with Nontradable Labor Income

This paper examines how risky labor income and retirement affect optimal portfolio choice. With idiosyncratic labor income risk, the optimal allocation to stocks is unambiguously larger for employed investors than for retired investors, consistent with the typical recommendations of investment advisors. Increasing idiosyncratic labor income risk raises investors' willingness to save and reduces their stock portfolio allocation towards the level of retired investors. Positive correlation between labor income and stock returns has a further negative effect and can actually reduce stockholdings below the level of retired investors.     

Dynamic portfolio selection with process control

The risk inherent in the accumulation of investment capital depends on the true return distributions of the risky assets, the accuracy of estimated returns, and the investment strategy. This paper considers risk control with Value-at-Risk and Conditional Value-at-Risk, using control limits to determine times for portfolio rebalancing. Optimal strategies and control limits are determined for a geometric Brownian motion asset pricing model with random parameters. The approaches to risk control are applied to the fundamental problem of investment in stocks, bonds, and cash over time.     

Time Diversification,Safety-First and Risk

The purpose of this article is to demonstrate the effect of investment time horizon on the choice of risky assets in a portfolio when the investor in question is optimizing a Safety-First (downside risk-aversion) utility function. It is shown, under standard assumptions, that although shortfall risk decreases exponentially with investment time horizon, the portfolio asset allocation proportions remain invariant. In fact, in some instances, the optimal allocation will not even depend on the drift of the underlying assets. Thus, we extend the classical results of Samuelson and Merton, derived under conventional utility assumptions, to an individual optimizing an A.D. Roy Safety-First objective; a discontinuous utility function that has been extolled as conforming to observed investor behaviour. A numerical example is provided.     

Optimal portfolio choice for investors with industry-specific labor income risks

We study optimal investment decisions for long-horizon investors with industry-specific labor income risks. We find that in addition to the volatility of labor income growth, the correlation between labor income and risky asset returns is another important factor that affects the optimal portfolio decisions and may provide a plausible explanation for the mixed empirical evidence of the relationship between labor income risk and portfolio holdings. Depending on its relative covariance with stock and bond returns, labor income may help resolve or deepen the asset allocation puzzle.     

A dynamic model of active portfolio management with benchmark orientation

This paper studies optimal dynamic portfolios for investors concerned with the performance of their portfolios relative to a benchmark. Assuming that asset returns follow a multi-linear factor model similar to the structure of Ross (1976) [Ross, S., 1976. The arbitrage theory of the capital asset pricing model. Journal of Economic Theory, 13, 342–360] and that portfolio managers adopt a mean tracking error analysis similar to that of Roll (1992) [Roll, R., 1992. A mean/variance analysis of tracking error. Journal of Portfolio Management, 18, 13–22], we develop a dynamic model of active portfolio management maximizing risk adjusted excess return over a selected benchmark. Unlike the case of constant proportional portfolios for standard utility maximization, our optimal portfolio policy is state dependent, being a function of time to investment horizon, the return on the benchmark portfolio, and the return on the investment portfolio. We define a dynamic performance measure which relates portfolio’s return to its risk sensitivity. Abnormal returns at each point in time are quantified as the difference between the realized and the model-fitted returns. Risk sensitivity is estimated through a dynamic matching that minimizes the total fitted error of portfolio returns. For illustration, we analyze eight representative mutual funds in the U.S. market and show how this model can be used in practice.     

The Role of Learning in Dynamic Portfolio Decisions

This paper analyzes the effect of uncertainty about the mean return on the risky asset on the portfolio decisions of an investor who has a long investment horizon. Building on the earlier work of Detemple (1986), Dothan and Feldman (1986), and Gennotte (1986), it is shown that the possibility of future learning about the mean return on the risky asset induces the investor to take a larger or smaller position in the risky asset than she would if there were no learning, the direction of the effect depending on whether the investor is more or less risk tolerant than the logarithmic investor whose portfolio decisions are unaffected by the possibility of future learning. Numerical calculations show that uncertainty about the mean return on the market portfolio has a significant effect on the portfolio decision of an investor with a 20 year horizon if her assessment of the market risk premium is based solely on the Ibbotson and Sinquefield (1995) data.     

Conditional portfolio allocation: Does aggregate market liquidity matter?

This paper investigates how aggregate liquidity influences optimal portfolio allocations across various US characteristic portfolios. We consider short-term allocation problems, with single and multiple risky assets, and use the nonparametric approach of Brandt (1999) to directly express optimal portfolio weights as functions of aggregate liquidity shocks. We find, first, that the effect of aggregate liquidity is positive and decreasing with the investment horizon. Second, at daily and weekly horizons, this effect is weaker on allocations in large stocks and gets stronger as we move toward small stocks, regardless of the other stock characteristics, suggesting that liquidity is the main concern of very short-term investors. Third, conditional allocations in risky assets decrease and exhibit shifts toward more liquid assets as aggregate liquidity worsens. Overall, conditioning on aggregate liquidity yields empirical results that are consistent with the so-called flight-to-safety and flight-to-liquidity episodes. Finally, we propose a simple tactical investment strategy and show how aggregate liquidity information can be exploited to enhance the out-of-sample performance of long-term strategies.     

The Role of Learning in Dynamic Portfolio Decisions

This paper analyzes the effect of uncertainty about the meanreturn on the risky asset on the portfolio decisions of an investorwho has a long investment horizon. Building on the earlier workof Detemple (1986), Dothan and Feldman (1986), and Gennotte(1986), it is shown that the possibility of future learningabout the mean return on the risky asset induces the investorto take a larger or smaller position in the risky asset thanshe would if there were no learning, the direction of the effectdepending on whether the investor is more or less risk tolerantthan the logarithmic investor whose portfolio decisions areunaffected by the possibility of future learning. Numericalcalculations show that uncertainty about the mean return onthe market portfolio has a significant effect on the portfoliodecision of an investor with a 20 year horizon if her assessmentof the market risk premium is based solely on the Ibbotson andSinquefield (1995) data.     

International asset allocation under regime switching, skew, and kurtosis preferences

This paper investigates the international asset allocation effectsof time-variations in higher-order moments of stock returnssuch as skewness and kurtosis. In the context of a four-momentInternational Capital Asset Pricing Model (ICAPM) specificationthat relates stock returns in five regions to returns on a globalmarket portfolio and allows for time-varying prices of covariance,co-skewness, and co-kurtosis risk, we find evidence of distinctbull and bear regimes. Ignoring such regimes, an unhedged USinvestor's optimal portfolio is strongly diversified internationally.The presence of regimes in the return distribution leads toa substantial increase in the investor's optimal holdings ofUS stocks, as does the introduction of skewness and kurtosispreferences.     

Options Trading and the CAPM

This article studies equilibrium asset pricing when agents facenonnegative wealth constraints. In the presence of these constraintsit is shown that options on the market portfolio are nonredundantsecurities and the economy's pricing kernel is a function ofboth the market portfolio and the nonredundant options. Thisimplies that the options should be useful for explaining riskyasset returns. To test the theory, a model is derived in whichthe expected excess return on any risky asset is linearly related(via a collection of betas) to the expected excess return onthe market portfolio and to the expected excess returns on thenonredundant options. The empirical results indicate that thereturns on traded index options are relevant for explainingthe returns on risky asset portfolios.     

Small caps in international equity portfolios: the effects of variance risk

We show that predictable covariances between means and variances of stock returns may have a first order effect on portfolio composition. In an international asset menu that includes both European and North American small capitalization equity indices, we find that a three-state, heteroskedastic regime switching VAR model is required to provide a good fit to weekly return data and to accurately predict the dynamics in the joint density of returns. As a result of the non-linear dynamic features revealed by the data, small cap portfolios become riskier in bear markets, i.e., display negative co-skewness with other stock indices. Because of this property, a power utility investor ought to hold a well-diversified portfolio, despite the high risk premium and Sharpe ratios offered by small capitalization stocks. On the contrary, small caps command large optimal weights when the investor ignores variance risk, by incorrectly assuming joint normality of returns.      

Portfolio allocation and the investment horizon: a multiscaling approach

A new approach is proposed for analysing portfolio allocation over various time scales. This new approach is based on wavelet analysis, which decomposes a given time series on a scale-by-scale basis. Empirical results indicate that, as the investment horizon lengthens, a greater weighting should be allocated to stocks. An explanation for this result is that the mean-reverting property of stock returns causes investors to perceive that stocks are less risky than bonds and T-bills at longer time scales compared with shorter time scales. When we include the effect of risk aversion, it is found that the higher the risk aversion, the less the Sharpe ratio, indicating that a more conservative investor prefers a smoother consumption stream.     

Predictability and diversification benefits of investing in commodity and currency futures

We re-examine diversification benefits of investing in commodities and currencies by considering a risk-averse investor with mean-variance preferences who exploits the possibility of predictable time variation in asset return means, variances, and covariances. We implement unconditional and conditional efficient portfolio strategies designed to exploit this predictability, together with more traditional and/or ad hoc ones yet hitherto relatively unexplored in this context (including the equally weighted, fixed weight, volatility timing, and reward-to-risk timing strategies). We find that, for all portfolio strategies, commodities and currencies do not improve the investment opportunity set of the investor with an existing portfolio of stocks, bonds and T-bills, and an investment horizon of one month. Our findings, which reverse the conclusions of previous studies that focus on static portfolio strategies, are robust across several performance metrics.     

What is the impact of wealth shocks on asset allocation?

I test the assumption of constant relative risk aversion using U.S. macroeconomic data and analyse the role of wealth shocks in generating transitory changes in asset portfolio composition. I show that the risky asset share exhibits cyclical behavior and it is significantly (and positively) affected by unexpected variation in wealth. Therefore, the empirical evidence suggests that risk aversion is counter-cyclical. I also find that the portfolio share of housing wealth falls when the agent is faced with a positive wealth shock, i.e. housing is a hedge against unfavorable wealth fluctuations. Finally, considering a variety of wealth definitions, the results show that: (i) wealth effects are stronger for direct holdings of risky assets than for indirect holdings, which highlights that investors do not typically trade some assets such as pension or mutual funds; (ii) although significant, wealth effects on asset allocation are mainly temporary as agents quickly rebalance the asset portfolio composition (i.e. there is weak evidence of inertia or slow adjustment in asset allocation); and (iii) changes in expected returns partially explain the variation in risky asset allocation.     

The bond–stock mix under time-varying interest rates and predictable stock returns

I investigate the allocation of wealth to cash, bonds, and stocks, along with the bond-to-stock ratio (BSR) when interest rates are time-varying and stock returns are predictable via the dividend-price ratio (DPR). The bond–stock mix and the BSR vary with the deviation of the current level of the DPR from its long-run mean and the correlations between all asset classes. The BSR may decrease over time, which contradicts both previously reported results on the matter as well as popular advice. Finally, I show that it is only at the investment horizon that the BSR is independent of risk aversion.     

Dynamic Consumption and Portfolio Choice with Stochastic Volatility in Incomplete Markets

This paper examines the optimal consumption and portfolio-choiceproblem of long-horizon investors who have access to a risklessasset with constant return and a risky asset ("stocks") withconstant expected return and time-varying precision—thereciprocal of volatility. Markets are incomplete, and investorshave recursive preferences defined over intermediate consumption.The paper obtains a solution to this problem which is exactfor investors with unit elasticity of intertemporal substitutionof consumption and approximate otherwise. The optimal portfoliodemand for stocks includes an intertemporal hedging componentthat is negative when investors have coefficients of relativerisk aversion larger than one, and the instantaneous correlationbetween volatility and stock returns is negative, as typicallyestimated from stock return data. Our estimates of the jointprocess for stock returns and precision (or volatility) usingU.S. data confirm this finding. But we also find that stockreturn volatility does not appear to be variable and persistentenough to generate large intertemporal hedging demands.     

The Variation of Economic Risk Premiums in Real Estate Returns

We examine the predictable components of returns on stocks, bonds, and real estate investment trusts (REITs). We employ a multiple-beta asset pricing model and find that there are varying degrees of predictability among stocks, bonds, and REITs. Furthermore, we find that most of the predictability of returns is associated with the economic variables employed in the asset pricing model. The stock market risk premium is highly important in capturing the predictable variation in stock portfolios, and the bond market risk premiums (term and risk structure of interest rates) are important in capturing the predictable variation in bond portfolios. For REITs, however, both the stock and bond market risk premiums capture the predictable variation in returns. REITs have comparable return predictability to stock portfolios. We conclude that there is an important economic risk premium for REITs that are not captured by traditional multiple-beta asset pricing models.