Campbell R. Harvey
Duke University, Durham, NC 27708, USA
National Bureau of Economic Research, Cambridge, MA 02138, USA
Christopher M. Kirby
University of Maryland, College Park, MD 20742, USA
Abstract
This paper outlinews a new procedure for constructing analytic tests of linear and non-linear factor pricing models. First, we derive a general representation of factor pricing models that is fully consistent with intertemporal utility maximization and that nests a number of well-known linear pricing results as special cases. Then we exploit this general representation to develop a straightforward approach for constructing analytic versions of the efficient generalized method of moments (GMM) estimator and the test for overidentifying restrictions for such models. No restrictions on serial correlation or conditional heteroskedasticity are required. All of our results are derived in a dynamic setting so the test assets can include dynamic trading strategies. Thus, our procedure for constructing analytic GMM tests is applicable to almost any factor pricing model that is likely to be encountered in practice.