Joint estimation of gradual variance changepoint for panel data with common structures

Abstract

This paper considers the problem of detecting the common gradual changepoint in panel data. Given the situation that each panel/series has the common gradual variance changepoint along with a smoothly functional mean, our focus is on improving the changepoint location estimation accuracy assuming a prior that a single change has occurred. We developed an iterative algorithm motivated by the idea of divide-and-conquer for tackling this problem. In addition, we have established the rates of convergence for all the parameter estimates. Monte Carlo simulation results illustrate the effectiveness and improvement of the proposed method over the existing methods. The method is also demonstrated through an analysis on the daily returns of S&P 500 Index and NASDAQ-100 Index over a period of time.