Testing the restrictions implied by the rational expectations hypothesis

The nature and form of the restrictions implied by the rational expectations hypothesis are examined in a variety of models with expectations and the properties of appropriate test statistics are analyzed with Monte Carlo evidence. Specifically, we consider the implications of lagged variables, simultaneous equations, and future period expectations upon the number and functional form of the rational expectations restrictions. Two asymptotically equivalent test statistics — a likelihood ratio and a Wald test — are available for implementing a test of these restrictions. Monte Carlo evidence is offered to provide a comparison between the properties of the alternative test statistics in small samples.     

The Effect of Errors in Variables on Tests for a Risk Premium in Forward Exchange Rates

Conventional tests for a risk premium in the price of forward exchange use the subsequently realized spot rate as a proxy for prior expectations. Use of this proxy creates a serious errors-in-variables problem which makes it difficult to reject the null hypothesis of zero risk premium. Use of a better proxy for expectations indicates the presence of a risk premium in the forward exchange rate of all countries analyzed.     

Labor market equilibrium under nonsequential search

This study presents a full model of the labor market under imperfect information. Workers bear the burden of search and are assumed to use a stochastic variant of an optimal sample size search strategy. The existence and uniqueness of a nondegenerate equilibrium distribution of wage offers is established. Infficiencies which lead to underinvestment in search are uncovered and analyzed.