Title: Quantile lasso with changepoints in panel data models
Authors: Matus Maciak - Charles University (Czech Republic) [presenting]
Abstract: Panel data models are quite modern statistical tools and they are commonly used in all kinds of econometric problems. In our approach we consider panel data models with changepoints, and atomic pursuit methods are utilized to detect and estimate these changepoints in the model. In order to obtain robust estimates and, also, to have a more complex insight into the data, we adopt the quantile lasso approach and the final model is obtained in a fully data-driven manner in just one modelling step. The main theoretical results are presented and some inferential tools for changepoint significance are proposed. The presented methodology is applied for a real data scenario and some finite sample properties are investigated via s simulation study.