B1222
Title: Sparse estimation in Markov regime-switching models
Authors: Gilberto Chavez Martinez - McGill University (Canada) [presenting]
Ankush Agarwal - University of Glasgow (United Kingdom)
Abbas Khalili - McGill University (Canada)
Ejaz Ahmed - Brock (Canada)
Abstract: Markov regime-switching vector auto-regression models are frequently used for modelling heterogeneous and complex relationships between variables in multivariate time series analysis. Applications include analyzing macroeconomic time series such as manufacturing activities, consumer price indices, and housing and asset prices. The most common estimation method in these models is maximum likelihood estimation (MLE). However, the MLE becomes unstable even for moderate data dimensions and some regimes. Regularization-based estimators are presented when the number of regimes in the model is correctly or over-specified. Theoretical and finite-sample performances of the methods are discussed, including forecasting, concluding with a real data analysis.
