On detecting jumps in time series: nonparametric setting

Motivated by applications in statistical quality control and signal analysis, we propose a sequential detection procedure which is designed to detect structural changes, in particular jumps, immediately. This is achieved by modifying a median filter by appropriate kernel-based jump-preserving weights (shrinking) and a clipping mechanism. We aim at both robustness and immediate detection of jumps. Whereas the median approach ensures robust smooths when there are no jumps, the modification ensure immediate reaction to jumps. For general clipping location estimators, we show that the procedure can detect jumps of certain heights with no delay, even when applied to Banach space-valued data. For shrinking medians, we provide an asymptotic upper bound for the normed delay. The finite sample properties are studied by simulations which show that our proposal outperforms classical procedures in certain respects.

[1]  Ing Rj Ser.  Approximation Theorems of Mathematical Statistics , 1980 .

[2]  A. Steland.  Sequential control of time series by functionals of kernal-weighted empirical processes under local alternatives , 2004 .

[3]  Jong-Sen Lee,et al.  Digital image smoothing and the sigma filter , 1983, Comput. Vis. Graph. Image Process..

[4]  W. Krumbholz,et al.  ASN-minimax double sampling plans for variables , 2002 .

[5]  M. Pollak.  Optimal Detection of a Change in Distribution , 1985 .

[6]  M. Pawlak,et al.  Jump preserving signal reconstruction using vertical weighting , 2001 .

[7]  J. Marron,et al.  Edge-Preserving Smoothers for Image Processing , 1998 .

[8]  B. E. Brodsky,et al.  Non-Parametric Statistical Diagnosis , 2000 .

[9]  C. Hesse.  A Bahadur-Type Representation for Empirical Quantiles of a Large Class of Stationary, Possibly Infinite-Variance, Linear Processes , 1990 .

[10]  D. G. Simpson,et al.  Edge-Preserving Smoothers for Image Processing: Comment , 1998 .

[11]  Adam Krzyzak,et al.  Clipped median and space-filling curves in image filtering , 2001 .

[12]  A. Vasilopoulos,et al.  Modification of Control Chart Limits in the Presence of Data Correlation , 1978 .

[13]  Ansgar Steland,et al.  Nonparametric monitoring of financial time series by jump-preserving control charts , 2002 .

[14]  B. Brodsky,et al.  Nonparametric Methods in Change Point Problems , 1993 .

[15]  Benjamin Yakir.  A note on optimal detection of a change in distribution , 1997 .

[16]  B. E. Brodsky,et al.  Non-Parametric Statistical Diagnosis: Problems and Methods , 2000 .

[17]  EWA SKUBALSKA-RAFAJ.  THE CLOSED CURVE FILLING MULTIDIMENSIONAL CUBE , 1994 .

[18]  B. Yakir.  Dynamic sampling policy for detecting a change in distribution, with a probability bound on false alarm , 1996 .

[19]  E. S. Page.  A test for a change in a parameter occurring at an unknown point , 1955 .

[20]  Benjamin Yakir.  A lower bound on the ARL to detection of a change with a probability constraint on false alarm , 1996 .

[21]  Y. Ritov.  Decision Theoretic Optimality of the Cusum Procedure , 1990 .

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

[23]  Fred Godtliebsen,et al.  M-Smoother with local Linear Fit , 2002 .

[24]  R. R. Bahadur.  A Note on Quantiles in Large Samples , 1966 .

[25]  S. W. Roberts,et al.  Control Chart Tests Based on Geometric Moving Averages , 2000, Technometrics.

[26]  G. Moustakides.  Optimal stopping times for detecting changes in distributions , 1986 .

[27]  J. Kiefer.  On Bahadur's Representation of Sample Quantiles , 1967 .

[28]  Wolfgang Schmid,et al.  Some properties of the EWMA control chart in the presence of autocorrelation , 1997 .

[29]  M. A. Girshick,et al.  A BAYES APPROACH TO A QUALITY CONTROL MODEL , 1952 .

[30]  A. Steland.  NP-optimal kernels for nonparametric sequential detection rules , 2004 .