Lingwei Kong

Publications

1. 2025

   Identification robust inference for the risk premium in term structure models

   With Frank Kleibergen

   Journal of Econometrics, 248, 105728.

   Article ↗arXiv

   Overview

   Identification robust tests for risk premia in dynamic affine term structure models. The paper develops inference that remains valid when risk factors are weakly identified and examines how weak identification affects estimates of time-varying risk prices.

2. 2023

   Identification Robust Testing of Risk Premia in Finite Samples

   With Frank Kleibergen and Zhaoguo Zhan

   Journal of Financial Econometrics, 21(2), 263–297.

   Paper PDFArticle ↗

   Abstract

   The reliability of tests on risk premia in linear factor models is threatened by limited sample sizes and weak identification of risk premia frequently encountered in applied work. We propose novel tests on the risk premia that are robust to both limited sample sizes and the identification strength of the risk premia as reflected by the quality of the risk factors. These tests are appealing for empirically relevant settings and lead to confidence sets of the risk premia that can substantially differ from conventional ones. To show the latter, we revisit two high-profile empirical applications.

3. 2023

   Rejoinder on: Identification Robust Testing of Risk Premia in Finite Samples

   With Frank Kleibergen and Zhaoguo Zhan

   Journal of Financial Econometrics, 21(2), 311–315.

   Article ↗

Working papers

1. 2026

   From dense grids to valid inference: Accounting for regularization bias in nonparametric random coefficient models

   With Maximilian Osterhaus and Michael Pen

   arXivPaper PDF

   Overview

   Inference for average functionals of nonparametrically estimated random-coefficient distributions. The proposed confidence intervals account for regularization bias, allowing dense grids to reduce approximation error. Applications include mean willingness-to-pay and average elasticities, with an illustration using travel mode demand.

2. 2026

   Transformer-based CoVaR: Systemic Risk in Textual Information (R&R, Journal of Financial Econometrics)

   With Junyu Chen, Tom Boot, and Weining Wang

   arXivPaper PDF

   Overview

   A Transformer estimator of Conditional Value-at-Risk that combines financial news embeddings with market data. The paper establishes error bounds and studies how textual information affects systemic-risk forecasts using U.S. returns and Reuters news.

3. 2025

   Plausible GMM: A Quasi-Bayesian Approach

   With Victor Chernozhukov, Christian B. Hansen, and Weining Wang

   Revised May 2026

   arXivPaper PDF

   Overview

   A quasi-Bayesian framework for structural inference when moment restrictions may be misspecified. Researchers express beliefs about misspecification through a prior distribution. The paper studies quasi-posterior concentration, Bayesian decision rules, and frequentist coverage, with empirical illustrations.

4. 2023

   Weak (Proxy) Factors Robust Hansen-Jagannathan Distance For Linear Asset Pricing Models

   arXivPaper PDF

   Abstract

   The Hansen-Jagannathan (HJ) distance statistic is one of the most dominant measures of model misspecification. However, the conventional HJ specification test procedure has poor finite sample performance, and we show that it can be size distorted even in large samples when (proxy) factors exhibit small correlations with asset returns. In other words, applied researchers are likely to falsely reject a model even when it is correctly specified. We provide two alternatives for the HJ statistic and two corresponding novel procedures for model specification tests, which are robust against the presence of weak (proxy) factors, and we also offer a novel robust risk premia estimator. Simulation exercises support our theory. Our empirical application documents the non-reliability of the traditional HJ test as it may produce counter-intuitive results when comparing nested models by rejecting a four-factor model but not the reduced three-factor model, while our proposed methods are practically more appealing and show support for a four-factor model for Fama French portfolios.