Abstract:Partially linear regression models are semiparametric models that contain both linear and nonlinear components. They are extensively used in many scientific fields for their flexibility and convenient interpretability. In such analyses, testing the significance of the regression coefficients in the linear component is typically a key focus. Under the high‐dimensional setting, i.e., “large p, small n,” the conventional F‐test strategy does not apply because the coefficients need to be estimated through regularization techniques. In this article, we develop a new test using a U‐statistic of order two, relying on a pseudo‐estimate of the nonlinear component from the classical kernel method. Using the martingale central limit theorem, we prove the asymptotic normality of the proposed test statistic under some regularity conditions. We further demonstrate our proposed test's finite‐sample performance by simulation studies and by analyzing some breast cancer gene expression data. Les modèles de régression partiellement linéaires sont des modèles semi‐paramétriques à deux composantes, l'une linéaire et l'autre non‐linéaire. De par leur flexibilité et leur facilité d'interprétation, ils sont largement utilisés dans de nombreux domaines scientifiques. En règle générale, dans de tels modèles, tester si des coefficients de la composante linéaire sont significatifs constitue un objectif clé. Aussi, dans le cadre de régression en haute dimension, c'est‐à‐dire «p est grand , n est petit», le test F usuel ne peut pas être utilisé car les coefficients du modèle sont estimés à l'aide de techniques de régularisation. Les auteurs de ce travail proposent un nouveau test basé sur une U‐statistique d'ordre deux reposant sur une pseudo estimation de la composante non linéaire. La pseudo estimation en question est construite à l'aide de l'estimateur à noyau classique. Ils obtiennent ensuite, sous certaines conditions de régularité, la normalité asymptotique de la statistique du test en faisant appel au théorème central limite des martingales. Enfin, ils illustrent la performance du test proposé sur des échantillons de taille finie à l'aide de simulations et présentent l'analyse d'un jeu de données génétiques sur le cancer du sein.

DIMENSION-REDUCTION TYPE TEST FOR LINEARITY OF A STOCHASTIC REGRESSION MODEL

Specification Testing in Nonstationary Time Series Models

Testing linearity in semi-functional partially linear regression models

LINEAR HYPOTHESIS TESTING FOR HIGH DIMENSIONAL GENERALIZED LINEAR MODELS.

Shrinkage-based regularization tests for high-dimensional data with application to gene set analysis

Heteroscedasticity Testing for Regression Models: A Dimension Reduction-based Model Adaptive

Consistent Specification Testing Via Nonparametric Series Regression

Approach to Testing Nonlinear Time Series Models

Diagnostic Checking for the Adequacy of Nonlinear Time Series Models

A General Approach to Testing Volatility Models in Time Series

A Joint Test for Parametric Specification and Independence in Nonlinear Regression Models

Linear Hypothesis Testing in Dense High-Dimensional Linear Models

Stability and Reliability Analysis of Nonlinear Stochastic System Using Data-Driven Dimensional Analysis Method

Linear Hypothesis Testing in Linear Models With High-Dimensional Responses

High-dimensional general linear hypothesis tests via non-linear spectral shrinkage

On A Robust Test for Setar-Type Nonlinearity in Time Series Analysis

A new test for high‐dimensional regression coefficients in partially linear models

Goodness-of-fit Testing in Linear Regression Models

Consistent Testing for Serial Correlation of Unknown Form

Nonlinearity and Spatial Lag Dependence: Tests Based on Double-Length Regressions

Tests for ultrahigh-dimensional partially linear regression models