We introduce an extension of finite mixture models by incorporating skewnormal distributions within a Hidden Markov Model framework assisted by a Viterbi algorithm. By estimating state-specific parameters, including location, scale, and skewness, the model enables accurate modelling of asymmetric data and detection of regime transitions, providing an alternative solution in cases where Gaussian mixtures may prove computationally ine!cient or prone to overfitting.
Skew-normal finite mixture models for change-point detection in time series / Forti, Marco; Nigri, Andrea; Shang, Hanlin. - (2025), pp. 1280-1286. ( 2025 Conference of the 12th Scientific Meeting of the Statistics for the Evaluation and Quality of Services Group of the Italian Statistical Society (SVQS) Bressanone ).
Skew-normal finite mixture models for change-point detection in time series
Marco Forti
;Andrea Nigri;
2025
Abstract
We introduce an extension of finite mixture models by incorporating skewnormal distributions within a Hidden Markov Model framework assisted by a Viterbi algorithm. By estimating state-specific parameters, including location, scale, and skewness, the model enables accurate modelling of asymmetric data and detection of regime transitions, providing an alternative solution in cases where Gaussian mixtures may prove computationally ine!cient or prone to overfitting.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
