Some Problems with Application of Change‐Point Detection Methods to Environmental Data

Change detection in dynamic attributed networks

“…In most scientific studies, data are generated as multivariate time series. Some examples include, climate studies where time varying data are collected on multiple variables such as temperature, precipitation and water discharges (Jarušková, ) and studies on financial markets where data on asset returns are observed over time (Lavielle & Teyssiere, ). Given multivariate time series data, a change in the correlation structure may indicate a change‐point in the overall system.…”

Inference for single and multiple change‐points in time series

“…Before we present the Bayes-type methodology for multivariate Gaussian series and regression models, we review of some of the major developments of this methodology in the literature. The method has been considered by several authors, including Kander and Zacks (1966), Gardner (1969), Sen and Srivastava (1973, 1975), MacNeill (1974, 1978a, 1978b, Jandhyala and MacNeill (1989, 1991, 1997, Nyblom (1989), Jandhyala (1993), Jandhyala and Minogue (1993), Tang and MacNeill (1993), MacNeill and Jandhyala (1993), Jandhyala and Jiang (1996), Xie and MacNeill (2006) and Jandhyala et al (2010). Kander and Zacks (1966) extended the one-sided statistic to the one-parameter exponential family.…”

An assessment of method for changepoint detection applied in tropospheric parameter time series given from numerical weather model.

“…This approach was recommended by Potter (1981) for detecting shifts in the mean of precipitation series. The same idea of finding inhomogeneities in the series { } i Y using the reference series { } i X was also discussed in the papers Alexandersson and Moberg (1997), Jarušková (1997) or Peterson et al (1998). A crucial book that covers statistical method used for the purposes pursued in this paper is Csörgő and Horváth (1997).…”