A method for robust estimation of dynamic mixtures of multivariate distributions is proposed. The EM algorithm is modified by replacing the classical M-step with high breakdown S-estimation of location and scatter, performed by using the bisquare multivariate S-estimator. Estimates are obtained by solving a system of estimating equations that are characterized by component specific sets of weights, based on robust Mahalanobis-type distances. Convergence of the resulting algorithm is proved and its finite sample behavior is investigated by means of a brief simulation study and n application to a multivariate time series of daily returns for seven stock markets.
Farcomeni, A., L., G. (2015). S-estimation of hidden Markov models. COMPUTATIONAL STATISTICS, 30, 57-80 [10.1007/s00180-014-0521-2].
S-estimation of hidden Markov models
FARCOMENI, Alessio;
2015-01-01
Abstract
A method for robust estimation of dynamic mixtures of multivariate distributions is proposed. The EM algorithm is modified by replacing the classical M-step with high breakdown S-estimation of location and scatter, performed by using the bisquare multivariate S-estimator. Estimates are obtained by solving a system of estimating equations that are characterized by component specific sets of weights, based on robust Mahalanobis-type distances. Convergence of the resulting algorithm is proved and its finite sample behavior is investigated by means of a brief simulation study and n application to a multivariate time series of daily returns for seven stock markets.File | Dimensione | Formato | |
---|---|---|---|
Farcomeni_S-estimation_2014.pdf
solo utenti autorizzati
Dimensione
511.78 kB
Formato
Adobe PDF
|
511.78 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Farcomeni_S-estimation-hidden_2015.pdf
solo utenti autorizzati
Dimensione
492.19 kB
Formato
Adobe PDF
|
492.19 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.