A Continuous Threshold Expectile Model

TL;DR

This paper introduces a continuous threshold expectile regression model that captures how covariate effects change smoothly around an unknown threshold, with efficient estimation and testing procedures supported by theoretical and simulation results.

Contribution

It develops a novel continuous threshold expectile regression framework with asymptotic properties, a computationally efficient threshold test, and demonstrates its effectiveness through simulations and real data applications.

Findings

Estimator for the threshold is root-n consistent.

The proposed test is computationally efficient and effective.

Simulation studies show good finite sample performance.

Abstract

Expectile regression is a useful tool for exploring the relation between the response and the explanatory variables beyond the conditional mean. This article develops a continuous threshold expectile regression for modeling data in which the effect of a covariate on the response variable is linear but varies below and above an unknown threshold in a continuous way. Based on a grid search approach, we obtain estimators for the threshold and the regression coefficients via an asymmetric least squares regression method. We derive the asymptotic properties for all the estimators and show that the estimator for the threshold achieves root-n consistency. We also develop a weighted CUSUM type test statistic for the existence of a threshold in a given expectile, and derive its asymptotic properties under both the null and the local alternative models. This test only requires fitting the model under the null hypothesis in the absence of a threshold, thus it is computationally more efficient than the likelihood-ratio type tests. Simulation studies show desirable finite sample performance in both homoscedastic and heteroscedastic cases. The application of our methods on a Dutch growth data and a baseball pitcher salary data reveals interesting insights.