Robert E. Cumby, John Huizinga, Testing the Autocorrelation Structure of Disturbances in Ordinary Least Squares and Instrumental Variables Regressions, Econometrica, Vol. 60, No. 1 (Jan., 1992), pp. 185-195

In this paper we examine the Henriksson-Merton test of market timing and its potential usefulness in evaluating investment advice. The paper proposes a natural extension of the test that is valid under more general assumptions about the distribution of asset returns. We show that the Henriksson-Merton test and its more general counterpart are special cases of standard tests of market rationality and efficiency. Both tests are applied to a group of foreign exchange advisory services.

In this paper, the authors examine the performance of a sample of fifteen U.S.-based internationally diversified mutual funds between 1982 and 1988. Two performance measures are used, the Jensen measure and the positive period weighting measure proposed by Mark Grinblatt and Sheridan Titman. They find no evidence that the funds, either individually or as a whole, provide investors with performance that surpasses that of a broad, international equity index over this sample period.

In this paper, we examine the performance of a sample of fifteen U.S.‐based internationally diversified mutual funds between 1982 and 1988. Two performance measures are used, the Jensen measure and the positive period weighting measure proposed by Grinblatt and Titman. We find no evidence that the funds, either individually or as a whole, provide investors with performance that surpasses that of a broad, international equity index over this sample period.

The fiscal theory of price determination suggests that if primary surpluses evolve independently of government debt, the equilibrium price level “jumps” to assure fiscal solvency. In this non-Ricardian regime, fiscal policy—not monetary policy—provides the nominal anchor. Alternatively, in a Ricardian regime, primary surpluses are expected to respond to debt in a way that assures fiscal solvency, and the price level is determined in conventional ways. This paper argues that Ricardian regimes are as theoretically plausible as non-Ricardian regimes, and provide a more plausible interpretation of certain aspects of the postwar U.S. data than do non-Ricardian regimes.

Standard approaches to designing a futures hedge often suffer from two major problems. First, they focus only on minimizing risk, so no account is taken of the impact on expected return. Second , in estima ting the hedge ratio, no allowance is made for time variation in the distribution of cash and futures price changes. This paper describes a technique for estimating the optimal futures hedge that corrects these problems and illustrates its use in hedging Treasury bonds with T-bond futures.