Abstract | ||
---|---|---|
This paper presents a new robust method to estimate the parameters of ARIMA models. This method makes use of the minimum Hellinger distance estimator (MHDE) to- gether with a robust filter cleaner able to reject a large frac- tion of outliers, and a Gaussian maximum likelihood esti- mation which handles missing values. The main advantages of the procedure are its easiness, robustness, high efficiency and practical execution. Its effectiveness is demonstrated on Monte Carlo simulations and an example of the forecasting of the French daily electricity consumptions. Index Terms- Robustness, time series, Hellinger dis- tance, ARIMA models, outliers, load forecasting. |
Year | DOI | Venue |
---|---|---|
2009 | 10.5281/zenodo.41646 | EUSIPCO |
Field | DocType | Citations |
Econometrics,Mathematical optimization,Monte Carlo method,Hellinger distance,Computer science,Outlier,Autoregressive integrated moving average,Robustness (computer science),Gaussian,Missing data,Estimator | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 1 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yacine Chakhchoukh | 1 | 103 | 7.18 |