Paper Image

Principal components estimation for high dimensional factor models with weak factor loadings

Published on:

8 February 2024

Primary Category:

Econometrics

Paper Authors:

Jungjun Choi,

Ming Yuan

Bullets

Key Details

Studies principal components estimator for high-dimensional factor models with weak factor loadings

Shows estimator maintains consistency and asymptotic normality for any scaling exponent between 0 and 1

Complements existing results requiring stronger factors

Combines traditional eigendecomposition with leave-one-out analysis to relax assumptions

AI generated summary

Principal components estimation for high dimensional factor models with weak factor loadings

This paper studies the principal components estimator for high-dimensional approximate factor models, where the factor loading matrix scales sublinearly with the number of cross-sectional units. The authors show the estimator maintains consistency and asymptotic normality even when factor loadings are weak, complementing existing results that require stronger factors. A key contribution is combining traditional eigendecomposition techniques with more recent leave-one-out analysis to relax assumptions on factor strength and noise independence.

Answers from this paper

Comments

No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up