A Mallows-Type Model Averaging Estimator for the Varying-Coefficient Partially Linear Model

• Published in: Journal of the American Statistical Association Vol. 114; no. 526; pp. 882 - 892
• Main Authors: Zhu, Rong, Wan, Alan T. K., Zhang, Xinyu, Zou, Guohua
• Format: Journal Article
• Language: English
• Published: Alexandria Taylor & Francis 03.04.2019; Taylor & Francis Group, LLC; Taylor & Francis Ltd
• Subjects: Americans; Asymptotic optimality; Computer simulation; Criteria; Data analysis; Heteroscedasticity; Linear analysis; linear models; Mallows criterion; Model averaging; Prominence; Regression analysis; Simulation; Statistical methods; Statistics; Theory and Methods; Uncertainty; Varying-coefficient partially linear model; weight
• ISSN: 0162-1459; 1537-274X; 1537-274X
• DOI: 10.1080/01621459.2018.1456936

In the last decade, significant theoretical advances have been made in the area of frequentist model averaging (FMA); however, the majority of this work has emphasized parametric model setups. This article considers FMA for the semiparametric varying-coefficient partially linear model (VCPLM), which has gained prominence to become an extensively used modeling tool in recent years. Within this context, we develop a Mallows-type criterion for assigning model weights and prove its asymptotic optimality. A simulation study and a real data analysis demonstrate that the FMA estimator that arises from this criterion is vastly preferred to information criterion score-based model selection and averaging estimators. Our analysis is complicated by the fact that the VCPLM is subject to uncertainty arising not only from the choice of covariates, but also whether the covariate should enter the parametric or nonparametric parts of the model. Supplementary materials for this article are available online.