LAMORE: A Stable, Scalable Approach to Latent Vector Autoregressive Modeling of Categorical Time Series

Yubin Park, Carlos Carvalho, Joydeep Ghosh
Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, PMLR 33:733-742, 2014.

Abstract

Latent vector autoregressive models for categorical time series have a wide range of potential applications from marketing research to healthcare analytics. However, a brute-force particle filter implementation of the Expectation-Maximization (EM) algorithm often fails to estimate the maximum likelihood parameters due to the Monte Carlo approximation of the E-step and multiple local optima of the log-likelihood function. This paper proposes two auxiliary techniques that help stabilize and calibrate the estimated parameters. These two techniques, namely \textitasymptotic mean regularization and \textitlow-resolution augmentation, do not require any additional parameter tuning, and can be implemented by modifying the brute-force EM algorithm. Experiments with simulated data show that the proposed techniques effectively stabilize the parameter estimation process. Also, experimental results using Medicare and MIMIC-II datasets illustrate various potential applications of the proposed model and methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v33-park14, title = {{LAMORE: A Stable, Scalable Approach to Latent Vector Autoregressive Modeling of Categorical Time Series}}, author = {Park, Yubin and Carvalho, Carlos and Ghosh, Joydeep}, booktitle = {Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics}, pages = {733--742}, year = {2014}, editor = {Kaski, Samuel and Corander, Jukka}, volume = {33}, series = {Proceedings of Machine Learning Research}, address = {Reykjavik, Iceland}, month = {22--25 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v33/park14.pdf}, url = {https://proceedings.mlr.press/v33/park14.html}, abstract = {Latent vector autoregressive models for categorical time series have a wide range of potential applications from marketing research to healthcare analytics. However, a brute-force particle filter implementation of the Expectation-Maximization (EM) algorithm often fails to estimate the maximum likelihood parameters due to the Monte Carlo approximation of the E-step and multiple local optima of the log-likelihood function. This paper proposes two auxiliary techniques that help stabilize and calibrate the estimated parameters. These two techniques, namely \textitasymptotic mean regularization and \textitlow-resolution augmentation, do not require any additional parameter tuning, and can be implemented by modifying the brute-force EM algorithm. Experiments with simulated data show that the proposed techniques effectively stabilize the parameter estimation process. Also, experimental results using Medicare and MIMIC-II datasets illustrate various potential applications of the proposed model and methods.} }
Endnote
%0 Conference Paper %T LAMORE: A Stable, Scalable Approach to Latent Vector Autoregressive Modeling of Categorical Time Series %A Yubin Park %A Carlos Carvalho %A Joydeep Ghosh %B Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2014 %E Samuel Kaski %E Jukka Corander %F pmlr-v33-park14 %I PMLR %P 733--742 %U https://proceedings.mlr.press/v33/park14.html %V 33 %X Latent vector autoregressive models for categorical time series have a wide range of potential applications from marketing research to healthcare analytics. However, a brute-force particle filter implementation of the Expectation-Maximization (EM) algorithm often fails to estimate the maximum likelihood parameters due to the Monte Carlo approximation of the E-step and multiple local optima of the log-likelihood function. This paper proposes two auxiliary techniques that help stabilize and calibrate the estimated parameters. These two techniques, namely \textitasymptotic mean regularization and \textitlow-resolution augmentation, do not require any additional parameter tuning, and can be implemented by modifying the brute-force EM algorithm. Experiments with simulated data show that the proposed techniques effectively stabilize the parameter estimation process. Also, experimental results using Medicare and MIMIC-II datasets illustrate various potential applications of the proposed model and methods.
RIS
TY - CPAPER TI - LAMORE: A Stable, Scalable Approach to Latent Vector Autoregressive Modeling of Categorical Time Series AU - Yubin Park AU - Carlos Carvalho AU - Joydeep Ghosh BT - Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics DA - 2014/04/02 ED - Samuel Kaski ED - Jukka Corander ID - pmlr-v33-park14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 33 SP - 733 EP - 742 L1 - http://proceedings.mlr.press/v33/park14.pdf UR - https://proceedings.mlr.press/v33/park14.html AB - Latent vector autoregressive models for categorical time series have a wide range of potential applications from marketing research to healthcare analytics. However, a brute-force particle filter implementation of the Expectation-Maximization (EM) algorithm often fails to estimate the maximum likelihood parameters due to the Monte Carlo approximation of the E-step and multiple local optima of the log-likelihood function. This paper proposes two auxiliary techniques that help stabilize and calibrate the estimated parameters. These two techniques, namely \textitasymptotic mean regularization and \textitlow-resolution augmentation, do not require any additional parameter tuning, and can be implemented by modifying the brute-force EM algorithm. Experiments with simulated data show that the proposed techniques effectively stabilize the parameter estimation process. Also, experimental results using Medicare and MIMIC-II datasets illustrate various potential applications of the proposed model and methods. ER -
APA
Park, Y., Carvalho, C. & Ghosh, J.. (2014). LAMORE: A Stable, Scalable Approach to Latent Vector Autoregressive Modeling of Categorical Time Series. Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 33:733-742 Available from https://proceedings.mlr.press/v33/park14.html.

Related Material