Novel method for structural change detection in time series using DP with an action cost

Even if an appropriate prediction expression and/or model is constructed to fit a time series, the model gradually begins to fail the prediction of the time series from some time point. In such a case, it is important not only to detect the failing situation quickly, but also to renew the prediction model as soon as possible after the detection. We formulate the structural change detection in a time series as an optimal stopping problem, using the concept of DP (dynamic programming). The cost function is defined as the sum of a loss cost by failing and an action cost after the detection. We propose a method for optimal solution and show the correctness. Also, we clarify the effectiveness by a numerical experiment.

[1]  Peter Willett,et al.  Some methods to evaluate the performance of Page's test as used to detect transient signals , 1999, IEEE Trans. Signal Process..

[2]  Irving S. Reed,et al.  On the sequential detection of emerging targets , 1965, IEEE Trans. Inf. Theory.

[3]  Alan S. Willsky,et al.  A survey of design methods for failure detection in dynamic systems , 1976, Autom..

[4]  S. Blostein,et al.  Quickest Detection of an Abrupt Change in a Random Sequence with Finite Change-Time , 1993, Proceedings. IEEE International Symposium on Information Theory.

[5]  R. Oaxaca.  Another Look at Tests of Equality between Sets of Coefficients in Two Linear Regressions , 1974 .

[6]  Venugopal V. Veeravalli,et al.  Asymptotics of quickest change detection procedures under a Bayesian criterion , 2002, Proceedings of the IEEE Information Theory Workshop.

[7]  Michèle Basseville,et al.  Detecting changes in signals and systems - A survey , 1988, Autom..

[8]  Denis Kouame,et al.  Effect of the compensation in abrupt model change detection problem , 1996, 1996 IEEE Digital Signal Processing Workshop Proceedings.

[9]  A. K. Nandi,et al.  Multiresolution and hybrid Bayesian algorithms for automatic detection of change points , 1998 .

[10]  Richard F. Gunst,et al.  Applied Regression Analysis , 1999, Technometrics.

[11]  Rittwik Jana,et al.  Change detection in teletraffic models , 2000, IEEE Trans. Signal Process..

[12]  E. S. Page.  CONTINUOUS INSPECTION SCHEMES , 1954 .

[13]  Steven D. Blostein.  Quickest detection of a time-varying change in distribution , 1991, IEEE Trans. Inf. Theory.

[14]  Renaud J. Di Francesco.  Real-time speech segmentation using pitch and convexity jump models: application to variable rate speech coding , 1990, IEEE Trans. Acoust. Speech Signal Process..