In this paper I derive a risk-adjusted measure of portfolio performance (now known as "Jensen's Alpha") that estimates how much a manager's forecasting ability contributes to the fund's returns. The measure is based on the theory of the pricing of capital assets by Sharpe (1964), Lintner (1965a) and Treynor (Undated). I apply the measure to estimate the predictive ability of 115 mutual fund managers in the period 1945-1964—that is their ability to earn returns which are higher than those we would expect given the level of risk of each of the portfolios. The foundations of the model and the properties of the performance measure suggested here are discussed in Section II. The evidence on mutual fund performance indicates not only that these 115 mutual funds were on average not able to predict security prices well enough to outperform a buy-the-marketand-hold policy, but also that there is very little evidence that any individual fund was able to do significantly better than that which we expected from mere random chance. It is also important to note that these conclusions hold even when we measure the fund returns gross of management expenses (that is assume their bookkeeping, research, and other expenses except brokerage commissions were obtained free). Thus on average the funds apparently were not quite successful enough in their trading activities to recoup even their brokerage expenses. Keywords: Jensen's Alpha, mutual fund performance, risk-adjusted returns, forecasting ability, predictive ability.

This paper investigates nonlinear pricing kernels in which the risk factor is endogenously determined and preferences restrict the definition of the pricing kernel. These kernels potentially generate the empirical performance of nonlinear and multifactor models, while maintaining empirical power and avoiding ad hoc specifications of factors or functional form. Our test results indicate that preferencerestricted nonlinear pricing kernels are both admissible for the cross section of returns and are able to significantly improve upon linear single- and multifactor kernels. Further, the nonlinearities in the pricing kernel drive out the importance of the factors in the linear multi-factor model. A PRINCIPAL IMPLICATION OF THE Capital Asset Pricing Model ~CAPM! is that the pricing kernel is linear in a single factor, the portfolio of aggregate wealth. Numerous studies over the past two decades have documented violations of this restriction. 1 In response, researchers have examined the performance of alternative models of asset prices. These models have generally fallen into two classes: ~1! multifactor models such as Ross ’ APT or Merton’s ICAPM, in which factors in addition to the market return determine asset prices; or ~2! nonparametric models, such as Bansal et al. ~1993!, Bansal and Viswanathan ~1993!, and Chapman ~1997!, in which the pricing kernel is not

Abstract. Factor model is a very useful and popular model in finance. In this paper, we show the relation between factor model and blind source separation, and we propose to use Independent Component Analysis (ICA) as a data mining tool to construct the underlying factors and hence obtain the corresponding sensitivities for the factor model. 1

We construct portfolios of stocks and of bonds that are maximally predictable with respect to a set of ex ante observable economic variables, and show that these levels of predictability are statistically significant, even after controlling for data-snooping biases. We disaggregate the sources for predictability by using several asset groups, including size-sorted and industry-sorted portfolios, and find that the sources of maximal predictability shift considerably across sectors and size classes as the return-horizon changes. Using three out-of-sample measures of predictability, we show that the predictability of the maximally predictable portfolio is genuine and economically significant.

The paper deals with the issues associated with identification of stocks generating abnormal returns. Following the findings of a finance theory regarding portfolio tilting, a set of price-related stocks' attributes was analyzed. The analysis was conducted with the help of rough sets methodology which allows to distinguish "important" attributes for problem description, and to generate decision rules which can be later used to predict stocks' performance. Validity of the approach was tested on the Toronto Stock Exchange data. Keywords: rough sets, decision rules, reducts, portfolio tilting, anomalies theory About the Authors Roman Slowinski is Professor of Decision and Computer Sciences and Head of the Laboratory of Intelligent Decision Support Systems, Institute of Computer Science, Poznan University of Technology, Poznan, Poland. Robert Susmaga is on the research and teaching staff of the Institute of Computer Science at Poznan University of Technology, Poznan, Poland. Wojtek Mic...

Using both industry-specific data on 55 US industry sectors and an extensive range of macroeconomic variables, the authors compare the performance of nearest neighbour algorithms, OLS, and a number of two-stage models based on these two methods, when forecasting industry returns. As industry returns are a relatively under-researched area in the Finance literature, we also give a brief review of the existing theories as part motivation for our specific choice of variables, which are commonly employed by asset managers in practice. Performance is measured by the Information Coefficient (IC), which is defined as the average correlation between the 55 forecasted returns and the realised returns across industries over time. Due to transaction costs, investors and asset managers typically want a steady outperformance over time. Hence, the volatility of IC is taken into account through the application of “Sharpe Ratios”. We find that two-stage procedures mixing industry-specific information with macroeconomic indicators generally outperform both the stand-alone nearest neighbour algorithms and time-series based OLS macroeconomic models.

In this paper, we survey ideas and principles of modeling the investment decision process of economic agents. We start with the criteria of Markowitz of formulating return and risk as mean and variance, and also its extensions. We then look into other related criteria which are based on probability assumptions on future prices of securities. We also present methodologies which, instead of assuming probability distributions, rely on the best solution for the worst case scenario or in the average. A few multiple stage optimization models are discussed. Finally we give a few remarks on some interesting topics for further investigations.

By means of a detailed analysis of the returns of the Standard & Poors 500 (S&P 500) composite stock index over the last fifty years we show how theoretical results and methodological recommendations from the statistical theory of non-parametric curve inference allow one to consistently estimate expected return and volatility. In this approach we do not postulate an apriori relationship risk-return nor do we specify the evolution of the first two moments through covariates. Our analysis gives statistical evidence that the expected return of the S&P 500 index as well as the market price of risk (the ratio expected return minus risk free interest rate over volatility) vary significantly through time both in size and sign. In particular, the periods of negative (positive) estimated expected return and market price of risk coincide with the bear (bull) markets of the index as defined in the literature. A complex relationship between risk and expected return emerges which is far from the common assumption of a positive linear time-invariant relation.

Although the correlation between the public and private market pricing of real estate has generated considerable research effort, the methods utilized in previous studies have failed to capture the dynamic nature of this correlation. This paper proposes a new statistical method to address this issue. This method, known as the dynamic conditional correlation GARCH model, will enable us to study the dynamics of the correlation between the two markets over time and enrich our understanding of the public and private market pricing of real assets. We also differentiate among different real estate types. We find that the correlation between NAV returns and REIT returns is dynamic for all REIT types, except for the Office and Hotel REITs, and there is a strong degree of persistence in the series of correlation Our Granger-causality tests show that while Apartment, Hotel, Office, Self Storage and Diversified REIT prices lead their NAVs, the causality is not significant for the Industrial and Strip Center REITs. Furthermore, the causality is in the reverse direction for Mall REITs where NAVs lead REIT prices. We are grateful to Jim Clayton, Joseph Pagliari and Dogan Tirtiroglu for their helpful

Abstract. Markowitz’s portfolio selection problem chooses weights for stocks in a portfolio based on a covariance matrix of stock returns. Our study proposes to reduce noise in the estimated covariance matrix using a Tikhonov filter function. In addition, we propose a new strategy to resolve the rank deficiency of the covariance matrix, and a method to choose a Tikhonov parameter which determines a filtering intensity. We put the previous estimators into a common framework and compare their filtering functions for eigenvalues of the correlation matrix. Experiments using the daily return data of the most frequently traded stocks in NYSE, AMEX, and NASDAQ show that Tikhonov filtering estimates the covariance matrix better than methods of Sharpe who applies a market-index model, Ledoit et al. who shrink the sample covariance matrix to the market-index covariance matrix, Elton and Gruber, who suggest truncating the smallest eigenvalues, Bengtsson and Holst, who decrease small eigenvalues at a single rate, and Plerou et al. and Laloux et al., who use a random matrix approach. Key words. Tikhonov regularization, covariance matrix estimate, Markowitz portfolio selection, ridge regression 1. Introduction. A