Spring 2001 Seminar Series - February 27, 2001
University of Minnesota
School of Statistics
College of Liberal Arts

Change-point Problem

Wei Biao Wu
University of Michigan
(Statistics Search Candidate)

Tuesday, February 27, 2001
4:00 PM, 211 Vincent Hall
Minneapolis, East Bank Campus
Social at 3:30 PM, 300 Ford Hall

Abstract

Many time series can be modeled as the sum of three components: long-time trend, seasonal effect and background noise. The trend superimposed with the seasonal effect constitute the mean of the process. The issue of mean stationarity is usually the first step for further statistical inference. In this talk, we present a theory of testing and estimation for a monotonic trend and the identification of seasonal effects. Testing is cast as a generic change-point problem, or probabilistic diagnostics. The change-point problem has been one of the central issues of statistical inference for several decades. It includes, for example, testing for changes in weather patterns and disease rates. We are mainly concerned with a posteriori testing, using spectral analysis to determine periodic components and isotonic regression to estimate the trend. A distinctive feature of our approach is that these two problems can be treated simultaneously: isotonic regression gives estimators for long-time trend with negligible influence from seasonal effects.