Change Point Detection in Panel Linear Regression Models Based on Jump Information Criterion

This paper focuses on the critical issue of change point detection in panel linear regression models and proposes a novel jump information criterion (JIC) for efficient solution. The core innovation of this criterion lies in reconstructing the traditional change point hypothesis testing problem into a parameter estimation problem: under the null hypothesis (H0, i.e., no change point exists in the model) and the alternative hypothesis (H1, i.e., a change point exists in the model), the number of potential change points is set to 0 and 1 for modeling and solution, respectively. To verify the theoretical reliability of the proposed method, this paper systematically establishes the consistency of the change point count estimator through rigorous mathematical deductions and further derives its optimal convergence rate. In terms of numerical validation, extensive Monte Carlo simulation experiments and real data empirical analysis both demonstrate that the estimator constructed based on JIC exhibits excellent performance in change point identification accuracy, stability, and computational efficiency, providing a reliable new tool for structural break analysis in panel data models.