Title: MOSUM-based test and estimation method for multiple changes in panel data
Authors: Man Wang - Donghua University (China) [presenting]
Abstract: Common change point detection is a vibrant topic in panel data analysis, however, most existing tests are based on the cumulative sum (CUSUM) method and suffer power loss under certain multiple change points setting. To solve this problem, a moving sum (MOSUM) based test is proposed to detect the common breaks happened in panel data with cross sectional dependence. Under mild conditions, it is shown that the proposed test statistic converges to an extreme distribution of a Gaussian process under the null hypothesis and diverges to infinity under the alternative hypothesis. Numerical studies show that the proposed method outperforms existing CUSUM-based procedures under multiple change points setting, as well as the single change point case with small sample size. We also give an estimation method based on the proposed test and establish its consistency. Application to the US state-level personal income data is also demonstrated.