A Data-Cleaning Augmented Kalman Filter for Robust Estimation of State Space Models
| Autor: | Stefano Grassi, Martyna Marczak, Tommaso Proietti |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Statistics and Probability
Economics and Econometrics Engineering Computer science Monte Carlo method Context (language use) Machine learning computer.software_genre 01 natural sciences 010104 statistics & probability Extended Kalman filter 0502 economics and business DISTRIBUTIONS State space Structural time series model 0101 mathematics Time series Additive outlier 050205 econometrics Innovation outlier Series (mathematics) business.industry 05 social sciences Augmented Kalman filter Estimator Kalman filter Robust filtering SCORING RULES Filter (video) Settore SECS-S/03 - Statistica Economica Ensemble Kalman filter OUTLIER DETECTION Artificial intelligence Statistics Probability and Uncertainty business computer Algorithm |
| Zdroj: | Marczak, M, Proietti, T & Grassi, S 2018, ' A data-cleaning augmented Kalman filter for robust estimation of state space models ', Econometrics and Statistics, vol. 5, no. 1, pp. 107-123 . https://doi.org/10.1016/j.ecosta.2017.02.002 |
| ISSN: | 1556-5068 |
| DOI: | 10.2139/ssrn.2756074 |
| Popis: | This article presents a robust augmented Kalman filter that extends the data– cleaning filter (Masreliez and Martin, 1977) to the general state space model featuring nonstationary and regression effects. The robust filter shrinks the observations towards their one–step–ahead prediction based on the past, by bounding the effect of the information carried by a new observation according to an influence function. When maximum likelihood estimation is carried out on the replacement data, an M–type estimator is obtained. We investigate the performance of the robust AKF in two applications using as a modeling framework the basic structural time series model, a popular unobserved components model in the analysis of seasonal time series. First, a Monte Carlo experiment is conducted in order to evaluate the com- parative accuracy of the proposed method for estimating the variance parameters. Second, the method is applied in a forecasting context to a large set of European trade statistics series. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|