Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Rational expectations models in which the forecast horizon is not equal to the sampling interval may result in serially correlated errors. Valid inference requires adjustment to the variance-covariance matrix. This paper outlines SEREM (Statistical Estimation of Rational Expectations Models)--a computer package written in GAUSS that computes a number of estimators that have been proposed to circumvent this problem. An application of these estimators to U.K. consumption suggests that the intertemporal elasticity of substitution for consumption is numerically close to, but significantly different from, zero. Copyright 1993 by Blackwell Publishing Ltd