Friday, November 15, 2019, 2:30pm
Some recent results in change point analysis
Marie Huskowa (Charles University Prague)
The change point problem is usually treated by statistical procedures for detection of instabilities in statistical models. The problem is usually formulated in terms of hypothesis testing and estimation problem. Typically, we have observations X1, …, Xn obtained at ordered time points and the basic task is to decide whether the model remains stable during the whole observational period or whether the model changes at some unknown point(s) or become generally instable. In case of change(s) in the model being detected the further task is also to estimate the time of a change and other parameters of the model in the periods where the model is stable. Such problems are also called disorder problems or testing for presence of structural breaks (in econometrics) or testing for stability or segmented regression or switching regression in the regression setup. If all n observations are available at the beginning of the statistical analysis we speak about a retrospective setup. If observations are arriving sequentially and after each new observation we have to decide whether the observations obtained so far indicate an instability or not we have a sequential setup or monitoring. Originally such problems were studied within statistical quality control, however nowadays there are many applications in various areas, e.g. medical research, econometrics, financial models, risk management, environmetrics, climatology. It brings a number of interesting theoretical and computational problems.
Talk contains some theoretical results for various models including dependent, high-dimensional observations as well as functional type observations. Theoretical results will be accompanied by various applications.
Der Vortrag beginnt um 14.30 Uhr in Raum 008 (SeMath, Pontdriesch 14-16).
Alle Interessierten sind herzlich willkommen.
Ort: Raum 008/SeMath, Pontdriesch 14-16, 52062 Aachen