Abstract:The additive model and the varying-coefficient model are both powerful regression tools, with wide practical applications. However, our empirical study on a financial data has shown that both of these models have drawbacks when applied to locally stationary time series. For the analysis of functional data, Zhang and Wang have proposed a flexible regression method, called the varying-coefficient additive model (VCAM), and presented a two-step spline estimation method. Motivated by their approach, we adopt the VCAM to characterize the time-varying regression function in a locally stationary context. We propose a three-step spline estimation method and show its consistency and asymptotic normality. For the purpose of model diagnosis, we suggest an L-2-distance test statistic to check multiplicative assumption, and raise a two-stage penalty procedure to identify the additive terms and the varying-coefficient terms provided that the VCAM is applicable. We also present the asymptotic distribution of the proposed test statistics and demonstrate the consistency of the two-stage model identification procedure. Simulation studies investigating the finite-sample performance of the estimation and model diagnosis methods confirm the validity of our asymptotic theory. The financial data are also considered. Supplementary materials for this article are available online.

Identification of Non-Varying Coefficients in Varying-Coefficient Models

Estimation and Inference for Varying Coefficient Partially Nonlinear Models

Statistical inference in partially time-varying coefficient models

Variable selection in partially time-varying coefficient models

Varying Coefficient Partially Nonlinear Models with Nonstationary Regressors

A New Functional Estimation Procedure for Varying Coefficient Models

Testing the Constancy in Varying-Coefficient Regression Models

Estimation of a Semiparametric Varying-Coefficient Partially Linear Errors-in-variables Model

A Fuzzy Varying Coefficient Model and Its Estimation

Estimation and Inference for Varying Coefficient Partially Nonlinear Errors-in-variables Models.

Generalized varying coefficient partially linear measurement errors models

Partially functional linear varying coefficient model

Error Variance Estimation in Partially Linear Varying Coefficient Models

Tests for Variance Components in Varying Coefficient Mixed Models

Efficient Estimation of a Varying-Coefficient Partially Linear Binary Regression Model

Estimation of Varying-Coefficient Partially Linear Model with Measurement Error in the Parametric Part

Estimation and Identification of a Varying-Coefficient Additive Model for Locally Stationary Processes

Statistical Inference for Nonparametric Analysis of Covariance Via Varying Coefficient Models

Equivalence of Two Tests in Varying Coefficient Partially Linear Errors in Variable Model with Missing Responses

Efficient estimation of partially linear varying coefficient models

A Unified Variable Selection Approach for Varying Coefficient Models