Outlier Detection in Regression Models with ARIMA Errors using Robust Estimates
Authors:
- 17 December 2001
- journal article
- research article
- Published by Wiley in Journal of Forecasting
- Vol. 20 (8) , 565-579
Abstract
A diagnostic procedure for detecting additive and innovation outliers as well as level shifts in a regression model with ARIMA errors is introduced. The procedure is based on a robust estimate of the model parameters and on innovation residuals computed by means of robust filtering. A Monte Carlo study shows that, when there is a large proportion of outliers, this procedure is more powerful than the classical methods based on maximum likelihood type estimates and Kalman filtering. Copyright © 2001 John Wiley & Sons, Ltd.Keywords
Funding Information
- US Bureau of the Census (Joint Statistical Agreements Nos 90-55 and 91-34)
- Universidad de Buenos Aires (TW92)
- CONICET (PIP 4186/96)
- FONCyT (PICT 97 No. 03-00000-00576)
This publication has 18 references indexed in Scilit:
- Joint Estimation of Model Parameters and Outlier Effects in Time SeriesJournal of the American Statistical Association, 1993
- Forecasting time series with outliersJournal of Forecasting, 1993
- Measuring Influence in Dynamic Regression ModelsTechnometrics, 1991
- Outlier Detection and Time Series ModelingTechnometrics, 1989
- High Breakdown-Point Estimates of Regression by Means of the Minimization of an Efficient ScaleJournal of the American Statistical Association, 1988
- Estimation of Time Series Parameters in the Presence of OutliersTechnometrics, 1988
- Outliers, level shifts, and variance changes in time seriesJournal of Forecasting, 1988
- Influence Functionals for Time SeriesThe Annals of Statistics, 1986
- Robust Estimation of the First-Order Autoregressive ParameterJournal of the American Statistical Association, 1979
- Approximate conditional-mean type smoothers and interpolators1979