Nowcasting causality in mixed frequency vector autoregressive models

T.B. Götz, A.W. Hecq

Research output: Working paperProfessional

381 Downloads (Pure)

Abstract

This paper introduces the notion of nowcasting causality for mixed-frequency VARs as the mixed-frequency version of instantaneous causality. We analyze the relationship between nowcasting and Granger causality in the mixed-frequency VAR setting of Ghysels (2012) and illustrate that nowcasting causality can have a crucial impact on the significance of contemporaneous or lagged high-frequency variables in standard MIDAS regression models.
Original languageEnglish
Place of PublicationMaastricht
PublisherMaastricht University, Graduate School of Business and Economics
Publication statusPublished - 1 Jan 2013

Publication series

SeriesGSBE Research Memoranda
Number050

Cite this

Götz, T. B., & Hecq, A. W. (2013). Nowcasting causality in mixed frequency vector autoregressive models. Maastricht University, Graduate School of Business and Economics. GSBE Research Memoranda, No. 050