STA/BST 290 Seminar: Alexandra Piryatinska

STA 290 Seminar Series

Thursday, June 4, 2015, 4:10pm, MSB 1147 (Colloquium Room)

Refreshments at 3:30pm in MSB 4110 (Statistics Lounge)

Speaker:          Alexandra Piryatinska

(Dept. Mathematics, San Francisco State University)

Title:                Methodology for Segmentation of time series of arbitrary nature via $\epsilon$-complexity of continuous function its applications

Abstract:          In this talk  the problem of splitting    time series of arbitrary nature (stochastic,

deterministic, or mixed) into segments generated by the same mechanism will be  considered. The concept of $\epsilon$-complexity of continuous functions will be introduced. A  characterization of this quantity for H\"{o}lder continuous functions will be given.  On the basis of the $\epsilon$-complexity parameters, we propose a  technique for the segmentation of time series that does not require any prior   knowledge of how these series were generated.  The results of simulations, and applications to  EEG and stock market  data, demonstrate the efficiency of the proposed methodology.

(Joint work with Boris Darkhovsky).