Outlier Detections in Autoregressive Models

This article proposes a new technique for detecting outliers in autoregressive models and identifying the type as either innovation or additive. This technique can be used without knowledge of the true model order, outlier location, or outlier type. Specifically, we perturb an observation to obtain the perturbation size that minimizes the resulting residual sum of squares (SSE). The reduction in the SSE yields outlier detection and identification measures. In addition, the perturbation size can be used to gauge the magnitude of the outlier. Monte Carlo studies and empirical examples are presented to illustrate the performance of the proposed method as well as the impact of outliers on model selection and parameter estimation. We also obtain robust estimators and model selection criteria, which are shown in simulation studies to perform well when large outliers occur.

[1]  Bovas Abraham,et al.  A SCORE TEST FOR DETECTION OF TIME SERIES OUTLIERS , 1988 .

[2]  Daniel Peña,et al.  Detection of outlier patches in autoregressive time series , 1998 .

[3]  P. Rousseeuw.  Least Median of Squares Regression , 1984 .

[4]  S. Weisberg,et al.  Residuals and Influence in Regression , 1982 .

[5]  Howard Wainer,et al.  Robust Regression & Outlier Detection , 1988 .

[6]  Philip Hans Franses,et al.  Time Series Models for Business and Economic Forecasting , 1998 .

[7]  G. Ljung,et al.  On Outlier Detection in Time Series , 1993 .

[8]  R. Martin.  Robust Estimation for Time Series Autoregressions , 1979 .

[9]  M. Genton,et al.  Robust Simulation-Based Estimation of ARMA Models , 2001 .

[10]  G. C. Tiao,et al.  Estimation of time series parameters in the presence of outliers , 1988 .

[11]  C. Muirhead.  Distinguishing Outlier Types in Time Series , 1986 .

[12]  G. Box,et al.  Bayesian analysis of some outlier problems in time series , 1979 .

[13]  Irving Wingate Burr.  Statistical quality control methods , 1976 .

[14]  Peter J. Rousseeuw,et al.  Robust regression and outlier detection , 1987 .

[15]  A. McQuarrie,et al.  Regression and Time Series Model Selection , 1998 .

[16]  G. C. Tiao.  3 Autoregressive moving average models, intervention problems and outlier detection in time series , 1985 .

[17]  R. Tsay.  Outliers, Level Shifts, and Variance Changes in Time Series , 1988 .

[18]  Shizuhiko Nishisato,et al.  Elements of Dual Scaling: An Introduction To Practical Data Analysis , 1993 .

[19]  長 国強.  Outlier Detection for Stationary Time Series , 1998 .

[20]  B. Abraham,et al.  Outlier detection and time series modeling , 1989 .

[21]  R. Tsay.  Time Series Model Specification in the Presence of Outliers , 1986 .

[22]  Clifford M. Hurvich,et al.  Regression and time series model selection in small samples , 1989 .

[23]  David M. Allen,et al.  The Relationship Between Variable Selection and Data Agumentation and a Method for Prediction , 1974 .

[24]  D. Peña.  Influential Observations in Time Series , 1990 .

[25]  Alberto Luceñ,et al.  Detecting possibly non‐consecutive outliers in industrial time series , 1998 .

[26]  Eugene L. Grant,et al.  Statistical Quality Control , 1946 .

[27]  M. Otto,et al.  Outliers in Time Series , 1972 .

[28]  H. Akaike,et al.  Information Theory and an Extension of the Maximum Likelihood Principle , 1973 .

[29]  V. Yohai,et al.  Influence Functionals for Time Series , 1986 .

[30]  R. Martin,et al.  Leave‐K‐Out Diagnostics for Time Series , 1989 .