Debiased Inference for Dynamic Nonlinear Panels with Multi-dimensional Heterogeneities
We introduce a generic class of dynamic nonlinear heterogeneous parameter models that incorporate individual and time fixed effects in both the intercept and slope. These models are subject to the incidental parameter problem, in that the limiting distribution of the point estimator is not centered at zero, and that test statistics do not follow their standard asymptotic distributions as in the absence of the fixed effects. To address the problem, we develop an analytical bias correction procedure to construct a bias-corrected likelihood. The resulting estimator follows an asymptotic normal distribution with mean zero. Moreover, likelihood-based tests statistics -- including likelihood-ratio, Lagrange-multiplier, and Wald tests -- follow the limiting chi-squared distribution under the null hypothesis. Simulations demonstrate the effectiveness of the proposed correction method, and an empirical application on the labor force participation of single mothers underscores its practical importance.
Code (0)
등록된 구현이 없습니다.
Tasks
Vocal Bursts Valence PredictionMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Robust Estimation and Inference in Panels with Interactive Fixed Effects
We consider estimation and inference for a regression coefficient in panels with interactive fixed effects (i.e., with a factor structure). We demonstrate that existing estimators and confidence intervals (CIs) can be he…
validLeast squares estimation in nonstationary nonlinear cohort panels with learning from experience
We discuss techniques of estimation and inference for nonstationary nonlinear cohort panels with learning from experience, showing, inter alia, the consistency and asymptotic normality of the nonlinear least squares esti…
SurveyPenalized GMM Framework for Inference on Functionals of Nonparametric Instrumental Variable Estimators
This paper develops a penalized GMM (PGMM) framework for automatic debiased inference on functionals of nonparametric instrumental variable estimators. We derive convergence rates for the PGMM estimator and provide condi…
SLIM: Stochastic Learning and Inference in Overidentified Models
We propose SLIM (Stochastic Learning and Inference in overidentified Models), a scalable stochastic approximation framework for nonlinear GMM. SLIM forms iterative updates from independent mini-batches of moments and the…
Penalized Likelihood Inference with Survey Data
This paper extends three Lasso inferential methods, Debiased Lasso, $C(\alpha)$ and Selective Inference to a survey environment. We establish the asymptotic validity of the inference procedures in generalized linear mode…
Survey