B0337
Title: High-dimensional dynamic factor models with Markov switching
Authors: Erik Kole - Erasmus University Rotterdam (Netherlands) [presenting]
Christian Brownlees - UPF (Spain)
Abstract: Factor models have become the standard methodology used for forecasting in macro and finance. It is shown how standard dynamic factor models can be extended with Markov-switching. This general class of models can accommodate the breaks and instabilities that have been documented with regard to factor models applied to large panels of time series. Model properties are analyzed, such as conditional moments and stationarity based on an extensive canonical formulation of the model that makes the switching explicit. This formulation is used to relate the model to the general theory of factor models. Estimation is proposed based on conditional expectation maximization and forecasting techniques are proposed. In the empirical application, the out-of-sample benefits of dynamic factor models are shown with Markov-switching.
