Fast leave-one-out methods for inference, model selection, and diagnostic checking. (December 2020)

• Record Type: Journal Article
• Authors: Belotti, Federico; Peracchi, Franco
• Is Part Of: Stata journal. Volume 20:Number 4(2020)
• Page Start: 785
• Page End: 804
• Publication Date: 2020-12
• Subjects: st0617 -- jackknife2 -- jackknife -- heteroskedasticity-consistent standard errors -- cross-validation -- diagnostic checking -- predictive residuals; Statistics -- Periodicals; Statistics -- Computer programs -- Periodicals; 001.422
• DOI: 10.1177/1536867X20976312
• Languages: English
• ISSNs: 1536-867X

Abstract

In this article, we describejackknife2, a new prefix command for jackknifing linear estimators. It takes full advantage of the available leave-one-out formula, thereby allowing for substantial reduction in computing time. Of special note is thatjackknife2 allows the user to compute cross-validation and diagnostic measures that are currently not available afterivregress 2sls, xtreg, andxtivregress .