A0939
Title: Change point analysis with irregular signals: When did the COVID-19 pandemic start?
Authors: Wei Biao Wu - University of Chicago (United States) [presenting]
Abstract: The focus is on testing and estimating change points where signals after the change point can be highly irregular. The setting substantially differs from the existing literature, assuming signals are piecewise constant or vary smoothly. A two-step approach is proposed to estimate the location of the change point effectively in. The first step consists of a preliminary estimation of the change point that allows for obtaining unknown parameters in the second step. In the second step, a new procedure is used to determine the position of the change point. It is shown that the optimal OP(1) rate of convergence of the estimated change point to the true change point can be obtained. The method is applied to analyze health time series data and estimate 7 December 2019 as the starting date of the COVID-19 pandemic.
