LAUSR

This paper examines different beta estimation methodologies for a large set of Developed and Emerging international markets. For all markets, estimators based on daily data outperform those based on monthly data. The optimal window length is surprisingly homogeneous, at roughly 12 months for Developed Markets, while tending to be somewhat longer for Emerging Markets. The best estimators include a double-shrinkage, a long memory (FI), and a simple combination approach. The FI model generally yields the best predictions for both Developed and Emerging Markets. For portfolio formation, the double-shrinkage, FI, and combination estimators also perform best.