Outlier detection in regression models with ARIMA errors using robust estimates

Journal of Forecasting

Volumn 20, Issue 8, 2001, Pages 565-579

  Bianco, A M ^a   Garcia Ben M ^b   Martinez E J ^a   Yohai, V J ^b  

^a UNIVERSIDAD DE BUENOS AIRES   (Argentina)

^b Baez 478   (Argentina)

Author keywords

Additive outlier; Innovation outlier; Level shifts; Time series

Indexed keywords

ERROR ANALYSIS; MONTE CARLO METHODS; PARAMETER ESTIMATION; PERTURBATION TECHNIQUES; REGRESSION ANALYSIS; VECTORS;

ROBUST ESTIMATES;

TIME SERIES ANALYSIS;

EID: 0035678283     PISSN: 02776693     EISSN: None     Source Type: Journal    
DOI: 10.1002/for.768     Document Type: Article

References (25)

1. Expectation-maximization algorithms and the estimation of time series models in the presence of outliers
2. Outlier detection and time series modeling
3. Robust procedures for regression models with ARIMA errors
4. Leave-k-out diagnostics for time series (with discussion)
5. Effect of exogenous intervention on the estimation of time series parameters
6. Estimation of time series parameters in the presence of outliers
7. Forecasting time series with outliers
8. Joint estimation of the model parameters and outlier effects in time series
9. Robust estimation of the first-order autoregressive parameter
10. Outliers in time series
11. The maximum likelihood estimation of ARMA models by Kalman filtering
12. The effect of additive outliers on the forecasts from ARIMA models
13. Outliers in time series analysis. Some comments on their impact and their detection
14. On outlier detection in time series
15. Approximate conditional mean type smoothers and interpolators
16. Robust methods for ARIMA models
17. Influence functionals for time series
18. Approximate non-gaussian filtering with linear state and observation relations
19. Two issues in time series outlier detection using indicator variables
20. Measuring the importance of outliers in ARIMA models
21. Influential observations in time series
22. Measuring influence in dynamic regression models
23. Outliers, level shifts and variance changes in time series
24. High breakdown-point estimates of regression by means of the minimization of an efficient scale