2021/06/20 by Daniel Becker, Becker, Daniel, Aloïs Kneip +3
Mathematics · #Advanced Statistical Methods and Models #Econometrics (econ.EM) #FOS: Economics and business #Statistical Distribution Estimation and Applications #Statistical Methods and Inference
paper · pdf · doi:10.48550/arxiv.2106.10723
openalex publication_date 2021/06/20 · openalex created_date 2021/08/02 · openalex updated_date 2026/07/28
In this paper, a semiparametric partially linear model in the spirit of\nRobinson (1988) with Box- Cox transformed dependent variable is studied.\nTransformation regression models are widely used in applied econometrics to\navoid misspecification. In addition, a partially linear semiparametric model is\nan intermediate strategy that tries to balance advantages and disadvantages of\na fully parametric model and nonparametric models. A combination of\ntransformation and partially linear semiparametric model is, thus, a natural\nstrategy. The model parameters are estimated by a semiparametric extension of\nthe so called smooth minimum distance (SmoothMD) approach proposed by Lavergne\nand Patilea (2013). SmoothMD is suitable for models defined by conditional\nmoment conditions and allows the variance of the error terms to depend on the\ncovariates. In addition, here we allow for infinite-dimension nuisance\nparameters. The asymptotic behavior of the new SmoothMD estimator is studied\nunder general conditions and new inference methods are proposed. A simulation\nexperiment illustrates the performance of the methods for finite samples.\n