Testing the Traditional CCAPM in the US: A Revisit

Abstract

This study re-examines the traditional consumption-based capital asset pricing model (CCAPM) by largely updating US monthly samples of consumption and stock market return and following the methodology of generalized method of moments (GMM) with instrumental variables of Hansen and Singleton (1982). As a result, our investigations reveal the following facts for the US stock market. First, 1) in the cases of the CCAPM with consumption for nondurable goods and the CCAPM with consumption for nondurable goods and services, their discount rate parameters almost always take similar values that are slightly less than one. Next, 2) their risk aversion parameters more stably take small minus values in the CCAPM with consumption for nondurable goods than in the CCAPM with consumption for nondurable goods and services. Third, 3) in our empirical examinations, all estimated CCAPMs with two kinds of consumption are not rejected by the J-tests.