Global identification of the semiparametric Box–Cox model

Abstract

Abstract. This paper establishes the identifiability of the parameters of the BoxCox model under restrictions that do not require the disturbance in the model to be independent of the explanatory variables. The proposed restrictions are semiparametric in nature: they restrict the support of the conditional distribution of the disturbance but do not require the latter to be known. JEL Codes: C13, C14

Semi‐parametric estimation of non‐separable models: a minimum distance from independence approach

“…Note that we allow the support of the dependent variable Y to depend on the true value of , as in some well‐known examples of such as the Box–Cox transformation model (see e.g. Komunjer, 2009).…”