Title: Almost Uniform Bias Correction for Persistent Panel Local Projections

Speaker: Geert Dhaene

Time: 15:40-17:00, Friday, April 10, 2026

Venue: I-206, Boxue Building

Host: Institute for Advanced Economic Research

【Speaker】

Geert DHAENE is Professor (Emeritus) of Econometrics at KU Leuven. His research lies at the intersection of econometric theory and applied methodology, with a particular emphasis on panel data models, nonlinear estimation, and nuisance parameters. Professor Dhaene's contributions to econometrics are widely recognized. His publications appear in leading journals including Econometrica, The Review of Economic Studies, Journal of Econometrics, Econometric Theory, etc.

【Abstract】

Panel local projections are widely used for impulse response estimation in policy analysis, with the within-group (WG) estimator being the default estimator. This paper studies the properties of the WG estimator and two bias-corrected estimators: the analytical bias correction of Herbst and Johannson (2024) and the half-panel jackknife (HPJ) bias correction suggested by Mei, Sheng, and Shi (2026). Our setup considers an autoregressive control variable that is potentially persistent, with varying degrees of persistence. We show that the two bias corrections are not uniformly valid when the data are persistent. The Herbst and Johannson (2024) bias correction is not valid whenever the autoregressive parameter drifts sufficiently quickly to one. The HPJ bias correction is not valid when the autoregressive parameter is local-to-unity, but remains valid when it drifts to one more slowly or more quickly. In addition, we propose novel standard errors for the HPJ estimator that yield confidence intervals with much better coverage properties.

【More Info】

For more information of the seminar, scan the following QR code(s) to join Tencent QQ group (904 544 292) or WeChat group named "IAER Seminar (4)", please.

Written by: Wang Jie

Source: Institute for Advanced Economic Research