Picture child's feet next to pens, pencils and paper

Open Access research that is helping to improve educational outcomes for children

Strathprints makes available scholarly Open Access content by researchers in the School of Education, including those researching educational and social practices in curricular subjects. Research in this area seeks to understand the complex influences that increase curricula capacity and engagement by studying how curriculum practices relate to cultural, intellectual and social practices in and out of schools and nurseries.

Research at the School of Education also spans a number of other areas, including inclusive pedagogy, philosophy of education, health and wellbeing within health-related aspects of education (e.g. physical education and sport pedagogy, autism and technology, counselling education, and pedagogies for mental and emotional health), languages education, and other areas.

Explore Open Access education research. Or explore all of Strathclyde's Open Access research...

Bayesian compressed vector autoregressions

Koop, Gary and Korobilis, Dimitris and Pettenuzzo, Davide (2017) Bayesian compressed vector autoregressions. Journal of Econometrics. pp. 1-63. ISSN 0304-4076 (In Press)

[img] Text (Koop-etal-JE-2017-Bayesian-compressed-vector-autoregressions)
Koop_etal_JE_2017_Bayesian_compressed_vector_autoregressions.pdf
Accepted Author Manuscript
Restricted to Repository staff only until 12 April 2019.
License: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 logo

Download (1MB) | Request a copy from the Strathclyde author

Abstract

Macroeconomists are increasingly working with large Vector Autoregressions (VARs) where the number of parameters vastly exceeds the number of observations. Existing approaches either involve prior shrinkage or the use of factor methods. In this paper, we develop an alternative based on ideas from the compressed regression literature. It involves randomly compressing the explanatory variables prior to analysis. A huge dimensional problem is thus turned into a much smaller, more computationally tractable one. Bayesian model averaging can be done over various compressions, attaching greater weight to compressions which forecast well. In a macroeconomic application involving up to 129 variables, we find compressed VAR methods to forecast as well or better than either factor methods or large VAR methods involving prior shrinkage.