Title: Gradient boosting in generalized Markov-switching regression models
Authors: Timo Adam - Bielefeld University (Germany) [presenting]
Andreas Mayr - University of Bonn (Germany)
Thomas Kneib - University of Goettingen (Germany)
Roland Langrock - Bielefeld University (Germany)
Abstract: Markov-switching generalized additive models for location, scale and shape constitute a novel class of latent-state time series regression models that allow different state-dependent parameters of the response distribution - not only the mean, but also variance, skewness and kurtosis parameters - to be modelled as potentially smooth functions of a given set of explanatory variables. In addition, the set of possible distributions that can be specified for the response is not limited to the exponential family but additionally includes, for instance, a variety of Box-Cox-transformed, zero-inflated and mixture distributions. An estimation approach based on the EM algorithm is proposed, where the gradient boosting framework is used to prevent overfitting while simultaneously performing variable selection. The feasibility of the suggested approach is assessed in simulation experiments and illustrated in a real-data setting, where the conditional distribution of the daily average price of energy in Spain is modeled over time.