A1045
Title: Nowcasting GDP with factor-augmented high-dimensional MIDAS regression
Authors: Jonas Striaukas - Copenhagen Business School (Denmark) [presenting]
Abstract: A factor-augmented high-dimensional mixed-frequency regression model is introduced to nowcast the US GDP growth. The new approach builds on the literature of nowcasting using sparse methods and goes beyond it by combining sparse regression with factor models. The estimator's convergence rates in a time series context are derived, allowing for mixing processes and heavier than exponential tails. The new technique is applied to nowcast the US GDP growth, and among other insights, it is found that it significantly improves over a range of more traditional nowcasting methods, which are based on either sparse regression or factor models, but not both.