Stanley J. Kon

The Market-Timing Performance of Mutual Fund Managers

The theory and measurement of the investment performance of managed portfolios has received a great deal of attention in the financial economics literature. 1 The topic is important to potential clients, managers evaluating the effectiveness of their strategies, and regulators that formulate policy concerning the operations of the marketplace. The Jensen (1968, 1969) model of riskadjusted performance has been the premier methodology for addressing these issues and testing the strong form of the efficient markets hypothesis (that is, whether any market participant has monopolistic access to any information relevant for price formation). The investment management process can be dichotomized into the activities of stock selection and market timing. Stock selection is based on forecasts of company-specific events and hence the prices of individual securities. Market timing, however, refers to forecasts of future This paper proposes an empirical methodology for measuring the market-timing performance of an investment manager and provides evidence for a sample of mutual funds. The results indicate that at the individual fund level there is evidence of significant superior timing ability and performance. However, the multivariate tests were not inconsistent with the efficient markets hypothesis. That is, fund managers as a group have no special information regarding the formation of expectations on the returns of the market portfolio.

Alternative Models for the Conditional Heteroscedasticity of Stock Returns

This article compares econometric model specifications that have been proposed to explain the commonly observed characteristics of the unconditional distribution of daily stock returns. The empirical results indicate that the most likely ranking is (1) intertemporal dependence models, (2) Student t, (3) generalized mixture-of-normal distributions, (4) Poisson jump, and (5) the stationary normal. Among the intertemporal dependence models for conditional heteroscedasticity, those with a leverage (or asymmetry) effect are superior. The Glosten, Jagannathan, and Runkle specification is the most descriptive for individual stocks, while Nelsons exponential model is the most likely for stock indexes. Copyright 1994 by University of Chicago Press.

Structural change and time dependence in models of stock returns

Abstract In this paper, we provide evidence that the time-series properties of stock returns include both structural change and time dependence in the conditional variance. The absence of a structural change component tends to overstate the persistence parameter in a time-dependent model specification. The reason is that time-dependent model specifications do not distinguish between positive residuals that increase volatility from those that represent a resolution of uncertainty. A sequential mixture of normal distributions model of structural change is employed to estimate discrete change points in the time-series of volatility in either direction. Although the model of structural change appears to be the more descriptive process in a mutually exclusive comparison, a joint model of time dependence and structural change is most likely.

Cash demand, liquidation costs and capital market equilibrium under uncertainty

Abstract In this paper, the portfolio and the liquidity planning problems are unified and analyzed in one model. Stochastic cash demands have a significant impact on both the composition of an individuals optimal portfolio and the pricing of capital assets in market equilibrium. The derived capital asset pricing model with cash demands and liquidation costs shows that both the market price of risk and the systematic risk of an asset are affected by the aggregate cash demands and liquidity risk. The modified model does not require that all investors hold an identical risky portfolio as implied by the Sharpe-Lintner-Mossin model. Furthermore, it provides a possible explanation for the noted discrepancies between the empirical evidence and the prediction of the traditional capital asset pricing model.

The Value of Shelf Registration for New Debt Issues

The Securities and Exchange Commission (SEC) Rule 415 (shelf registration) allows eligible firms to register securities that they intend to issue in the next two years. Once the registration is declared effective, firms can enter the market instantaneously. Ineligible issuers, however, are required to notify the SEC at least forty-eight hours before entry. Supporters of the shelf rule contend that the rule provides valuable flexibility in timing issues. The authors examine this contention by comparing the market-timing performance of shelf and nonshelf debt issues. The results indicate that the shelf rule does provide valuable market-timing flexibility. Copyright 1989 by the University of Chicago.

Sequential Parameter Nonstationarity in Stock Market Returns

This paper provides a Bayesian test of parameter nonstationarity and an estimation procedure for the detection of change points in the time series of stock returns. The empirical results indicate that this procedure can identify the change points in the data without prior knowledge and provide substantially more descriptive validity for the distribution of stock returns than competing models.