Bruce N. Lehmann

Growth-Optimal Portfolio Restrictions On Asset Pricing Models

We show that absence of arbitrage in frictionless markets implies a lower bound on the average of the logarithm of the reciprocal of the stochastic discount factor implicit in asset pricing models. The greatest lower bound for a given asset menu is the average continuously compounded return on its growth-optimal portfolio. We use this bound to evaluate the plausibility of various parametric asset pricing models to characterize financial market puzzles such as the equity premium puzzle and the risk-free rate puzzle. We show that the insights offered by the growth-optimal bounds differ substantially from those obtained by other nonparametric bounds.

Performance measurement and evaluation

We consider performance measurement and evaluation for managed funds. Similarities and differences−both in econometric practice and in interpretation of outcomes of empirical tests−between performance measurement and conventional asset pricing models are analyzed. We also discuss how inference on ‘skill’ is affected when fund managers have market timing information. Performance testing based on portfolio weights is also covered as is recent developments in Bayesian models of performance measurement that can accommodate errors in the benchmark asset pricing model.

Earnings, Dividend Policy, and Present Value Relations: Building Blocks of Dividend Policy Invariant Cash Flows

In a Modigliani-Miller world, dividend policy is irrelevant for asset pricing. This article searches for cash flows with two characteristics: like dividends, asset prices can be calculated from their present values and, unlike dividends, they are invariant with respect to changes in dividend policy. Segmented and aggregate residual income measures with these features are identified under two assumptions: dividend policy does not alter risk premiums and income earned from investments associated with dividend policy includes unrealized capital gains and losses. The results hold for otherwise arbitrary risk premiums in the general no-arbitrage approach to the valuation of uncertain income streams.

Performance clustering and incentives in the UK pension fund industry

This paper analyses the performance of a large cross-section of UK pension funds. We ̄nd strong evidence of clustering in ex post average performance across pension fund portfolios as a whole as well as within asset classes. While there are robust di®erences in average performance in Jensen-style regressions from sources such as risk adjustment and the use of alternative peer-group and external benchmarks, the resulting cross-sectional variation in measured abnormal performance was virtually identical to that of the raw return data. Fund size appears to be the only variable that can account for an important fraction of the cross-sectional variation in measured performance. We explain these results in terms of the incentives operating in the UK pension fund industry. ¤We would like to thank Gordon Bagot and Val Ashmore of The WM Company for advice on interpreting the data used in this study.

Notes on Dynamic Factor Pricing Models

These notes discuss three aspects of dynamic factor pricing (i.e., APT) models. First, the diversifiable component of returns is unpredictable in a no-arbitrage world. Second, conditional factor loadings or betas have an unconditional factor structure when returns follow an unconditional factor structure, which provides a link between conditional and unconditional factor pricing models. Third, the estimation of dynamic factor pricing models is easily simplified in large cross sections when returns follow an unconditional factor structure. These results aid in the interpretation of existing applications and identify some of the issues in the formulation and estimation of dynamic factor pricing models.

Regularities in volatility and the price of risk following large stock market movements in the US and Japan

Abstract We examine the effects of large price changes on volatility and the price of risk of two broad stock indices, the CRSP value-weighted index of NYSE and AMEX stocks and the TOPIX of the Tokyo Stock Exchange. Volatility generally rises after large price changes of either sign but usually revert to pre-jump levels within a week or two. In contrast, the price of risk (except for the first day after a jump) typically declines following large price increases and rises following large price declines with little apparent tendency to revert toward pre-jump levels before the arrival of the next jump.

A multinomial characterization of feedforward neural networks

The purpose of the paper is to examine neural networks in terms of a particular probability model: a multinomial distribution characterization of the conditional mean. This characterization suggests circumstances in which networks need only provide good local approximations and a new parsimonious neural network model. The paper provides an empirical application to interest rate volatility.