Temporally-Reweighted Chinese Restaurant Process Mixtures for Clustering, Imputing, and Forecasting Multivariate Time Series
This method addresses the challenge of modeling hundreds of time series with non-stationary dynamics for applications like disease forecasting and economic analysis, representing an incremental improvement over existing techniques.
The authors tackled the problem of forecasting, imputation, and clustering in sparsely observed, multivariate time series by proposing a Bayesian nonparametric method, demonstrating superior forecasting accuracy and competitive imputation accuracy on seasonal flu data compared to baselines.
This article proposes a Bayesian nonparametric method for forecasting, imputation, and clustering in sparsely observed, multivariate time series data. The method is appropriate for jointly modeling hundreds of time series with widely varying, non-stationary dynamics. Given a collection of $N$ time series, the Bayesian model first partitions them into independent clusters using a Chinese restaurant process prior. Within a cluster, all time series are modeled jointly using a novel "temporally-reweighted" extension of the Chinese restaurant process mixture. Markov chain Monte Carlo techniques are used to obtain samples from the posterior distribution, which are then used to form predictive inferences. We apply the technique to challenging forecasting and imputation tasks using seasonal flu data from the US Center for Disease Control and Prevention, demonstrating superior forecasting accuracy and competitive imputation accuracy as compared to multiple widely used baselines. We further show that the model discovers interpretable clusters in datasets with hundreds of time series, using macroeconomic data from the Gapminder Foundation.