We propose a statistical model of differences in beliefs in which heterogeneous investors are represented as different machine learning model specifications. Each investor forms return forecasts from their own specific model using data inputs that are available to all investors. We measure...

We report experimental results for a twice-played prisoners' dilemma in which the players can choose the allocation of the stakes across the two periods. Our point of departure is the assumption that some (but not all) people are principled to "do the right thing," or cooperate, as long as their...

We theoretically characterize the behavior of machine learning asset pricing models. We prove that expected out-of-sample model performance--in terms of SDF Sharpe ratio and test asset pricing errors--is improving in model parameterization (or "complexity"). Our empirical findings verify the...

This paper examines the specification errors of several asset pricing models using the methodology of Hansen and Jagannathan (1997) and a common data set. The models are the CAPM, the Consumption CAPM, the Jagannathan and Wang (1996) conditional CAPM, the Campbell (1996) dynamic asset pricing...

The PPP puzzle is based on empirical evidence that international price differences for individual goods (LOOP) or baskets of goods (PPP) appear highly persistent or even non-stationary. The present consensus is these price differences have a half-life that is of the order of five years at best,...

From a sample of 910 U.S. firms over the period 1977 1996, we find that structure of the empirical model has significant impacts on resulting estimates of exchange rate exposures from equity returns. While lengthening the return horizon has minimal impact on exposure estimates, the inclusion of...

This paper examines monetary policy in Rudebusch and Svensson's (1999) two equation macroeconomic model when the policymaker recognizes that the model is an approximation and is uncertain about the quality of that approximation. It is argued that the minimax approach of robust control provides a...

A companion paper (Nelson (1992)) showed that in data observed at high frequencies, an ARCH model may do a good job at estimating conditional variances, even when the ARCH model is severely misspecified. While such models may perform reasonably well at filtering (i.e., at estimating unobserved...