Published online by Cambridge University Press: 27 October 2014
This paper investigates a general semiparametric stochastic discount factor formulation that avoids functional form misspecification. A new semiparametric estimation procedure is proposed which combines orthogonality conditions and local linear fitting to give a semiparametric generalized estimating equation approach. Asymptotic properties of the estimators are established and we explore the empirical usefulness of the proposed approach to value-weighted stock returns.
The authors would like to thank the editor, Professor Peter C.B. Phillips, the co-editor, Professor Oliver Linton, and the anonymous referees for their comments. Cai’s research was supported, in part, by the National Nature Science Foundation of China Grants #71131008 (Key Project), #70871003, and #70971113. Ren’s research was supported by a Natural Science Foundation of Fujian Province Grant (#2011J01384), the Fundamental Research Funds for the Central Universities (#2013221022), and the Natural Science Foundation of China Grants (#71301135, #71203189, #71131008).