Improved Bootstrap Inference for Dynamic Panel Data models with Interactive Effects

2026-09-25

AI Summary

This study addresses dynamic panel models with interactive effects, where conventional bootstrap methods fail to accurately capture bias and factor estimation variance under weak exogeneity. It elucidates the mechanism by which fixed regressors invalidate standard bootstrap procedures and proposes a recursive-design wild bootstrap inference method integrated with bias correction. By combining common correlated effects estimation with Monte Carlo simulation techniques, this approach innovatively unifies bootstrap resampling with existing bias-correction strategies to enhance statistical inference. Monte Carlo experiments demonstrate that the proposed method significantly outperforms conventional bias correction paired with cross-sectional bootstrap in terms of inferential accuracy.

Abstract

We study recursive-design wild bootstrap inference for dynamic panel data models with unobserved common factors estimated by Common Correlated Effects. In the large N,T setting, the bootstrap reproduces the biased limiting distribution in pure autoregressive models, but fails to capture all bias and factor-estimation variance components in models with additional regressors, particularly under weak exogeneity. We trace this failure to holding regressors fixed across bootstrap replications. We propose to combine bootstrap procedure with available bias-correction methods to conduct adjusted inference. Monte Carlo evidence shows substantial improvements over conventional strategies of using bias-correction paired with cross-sectional bootstrap methods.

Problem

• Dynamic Panel Data Models
• Bootstrap Inference
• Interactive Effects
• Common Correlated Effects
• Bias Correction

Innovation

• Dynamic Panel Data Models
• Wild Bootstrap
• Common Correlated Effects
• Bias Correction
• Interactive Effects