ANOVA-HDFE: Fast Variance Decomposition with High-Dimensional Fixed Effects and an Application to Trade Flows

Hanna L. Adam, Mario Larch and Michael Nower

No 12055, CESifo Working Paper Series from CESifo

Abstract: Performing an analysis of variance (ANOVA) on a large dataset spanning many dimensions becomes computationally challenging or even infeasible. We develop a new, fast procedure, ANOVA-HDFE, which uses sequential linear regressions and builds on recent advances in regression analysis with high-dimensional fixed effects (HDFE). It accommodates both balanced and unbalanced settings with many categorical and continuous covariates, while also allowing for high-dimensional fixed effects. Applying ANOVA-HDFE to bilateral trade flows, we find that 60% of the variation is at the country or country-time level. Moreover, a substantial proportion of the pair-specific variation remains unexplained by standard trade cost proxy variables.

Keywords: analysis of variance; high-dimensional fixed effects; large data; variation in high dimensions; variation of bilateral trade flows; asymmetric trade costs; ANOVA-HDFE (search for similar items in EconPapers)
JEL-codes: C23 C55 F14 F16 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.ifo.de/DocDL/cesifo1_wp12055.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ces:ceswps:_12055

Access Statistics for this paper

More papers in CESifo Working Paper Series from CESifo Contact information at EDIRC.
Bibliographic data for series maintained by Klaus Wohlrabe ().