Home

Factor Analysis Regression for Predictive Modeling with High-Dimensional Data

Download Link
https://econpapers.repec.org/scripts/redir.pf?u=http%3A%2F%2Flink.springer.com%2F10.1007%2Fs40953-022-00322-x;h=repec:spr:jqecon:v:20:y:2022:i:1:d:10.1007_s40953-022-00322-x
Time Added
2022/09/26 18:29
Search
Factor Model
Total Downloads
0
Authors
Randy Carter and Netsanet Michael Randy Carter: State University of New York at Buffalo Netsanet Michael: The Boeing Company
Abstract
Abstract Factor analysis regression (FAR) of $$y _i$$ y i on $${{\varvec{x}}}_i=(x _{1i}x _{2i}\ldots x _{pi})$$ x i = ( x 1 i x 2 i … x pi ) i = 12...n has been studied only in the low-dimensional case $$(p n )$$ ( p > n ) . In this paper we develop a high-dimensional version of FAR based on a computationally efficient method of factor extraction. We compare the performance of our high-dimensional FAR with partial least squares regression (PLSR) and principal component regression (PCR) under three underlying correlation structures: arbitrary correlation factor model correlation structure and when y is independent of x. Under each structure we generated Monte Carlo training samples of sizes $$n
Keywords
Bilinear factor model ; Principal component analysis ; Principal component regression ; Partial least squares ; Factor structure covariance matrix ; Factor analysis regression ; Mean square error of prediction ; Monte Carlo studies ; Cross-validation (search for similar items in EconPapers)
Year Published
2022
Series
Journal of Quantitative Economics 2022 vol. 20 issue 1 No 7 115-132
Rank
0.74
TOP