ABSTRACT
In econometrics and finance, variables are collected at different frequencies. One straightforward regression model is to aggregate the higher frequency variable to match the lower frequency with a fixed weight function. However, aggregation with fixed weight functions may overlook useful information in the higher frequency variable. On the other hand, keeping all higher frequencies may result in overly complicated models. In literature, mixed data sampling (MIDAS) regression models have been proposed to balance between the two. In this article, a new model specification test is proposed that can help decide between the simple aggregation and the MIDAS model.
Acknowledgments
The authors thank Dr. Sìlvia Gonçalves and the participants of 26th Annual Meeting of the Midwest Econometrics Group for constructive comments and suggestions. Superior, a high-performance computing cluster at Michigan Technological University, was used in obtaining results presented in this publication.