Identification without assuming mean stationarity: Quasi ML estimation of dynamic panel models with endogenous regressors

2020 ◽  
Author(s):  
Hugo Kruiniger
Keyword(s):  
Fixed Effects ◽  
Control Function ◽  
Limited Information ◽  
Dynamic Panel ◽  
Ml Estimation ◽  
Panel Models ◽  
Endogenous Regressors ◽  
Model Residuals ◽  
Stationarity Conditions ◽  
Dynamic Panel Models

Abstract Linear GMM estimators for dynamic panel models with predetermined or endogenous regressors suffer from a weak instruments problem when the data are highly persistent. In this paper we propose new random and fixed effects Limited Information Quasi ML estimators (LIQMLEs) for such models. We also discuss LIQMLEs for models that contain time-varying individual effects. Unlike System GMM estimators, the LIQMLEs do not require mean stationarity conditions for consistency. Such conditions often do not hold for the models we consider. Our LIQMLEs are based on a two-step control function approach that includes the first stage model residuals for a predetermined or endogenous regressor in the outcome equation. The LIMLEs are more precise than non-linear GMM estimators that are based on the original outcome equation. The LIQMLEs also compare favourably to various alternative (Q)MLEs in terms of precision, robustness and/or ease of computation.

scholarly journals Does job promotion affect men’s and women’s health differently? Dynamic panel models with fixed effects

2016 ◽  
pp. dyw310 ◽  
Author(s):  
Anna Nyberg ◽  
Paraskevi Peristera ◽  
Hugo Westerlund ◽  
Gunn Johansson ◽  
Linda L Magnusson Hanson
Keyword(s):  
Women’S Health ◽  
Women's Health ◽  
Fixed Effects ◽  
Dynamic Panel ◽  
Panel Models ◽  
Job Promotion ◽  
Dynamic Panel Models

scholarly journals Bidirectional Associations Between Depressive and Anxiety Symptoms and Loneliness During the COVID-19 Pandemic: Dynamic Panel Models With Fixed Effects

2021 ◽  
Vol 12 ◽  
Author(s):  
Cillian P. McDowell ◽  
Jacob D. Meyer ◽  
Daniel W. Russell ◽  
Cassandra Sue Brower ◽  
Jeni Lansing ◽  
...  
Keyword(s):  
Mental Health ◽  
Depressive Symptoms ◽  
Fixed Effects ◽  
Online Survey ◽  
Retention Rate ◽  
Anxiety Symptoms ◽  
Dynamic Panel ◽  
Panel Models ◽  
Bidirectional Associations ◽  
Dynamic Panel Models

Background: Understanding the direction and magnitude of mental health-loneliness associations across time is important to understand how best to prevent and treat mental health and loneliness. This study used weekly data collected over 8 weeks throughout the COVID-19 pandemic to expand previous findings and using dynamic panel models with fixed effects which account for all time-invariant confounding and reverse causation.Methods: Prospective data on a convenience and snowball sample from all 50 US states and the District of Colombia (n = 2,361 with ≥2 responses, 63.8% female; 76% retention rate) were collected weekly via online survey at nine consecutive timepoints (April 3–June 3, 2020). Anxiety and depressive symptoms and loneliness were assessed at each timepoint and participants reported the COVID-19 containment strategies they were following. Dynamic panel models with fixed effects examined bidirectional associations between anxiety and depressive symptoms and loneliness, and associations of COVID-19 containment strategies with these outcomes.Results: Depressive symptoms were associated with small increases in both anxiety symptoms (β = 0.065, 95% CI = 0.022–0.109; p = 0.004) and loneliness (β = 0.019, 0.008–0.030; p = 0.001) at the subsequent timepoint. Anxiety symptoms were associated with a small subsequent increase in loneliness (β = 0.014, 0.003–0.025; p = 0.015) but not depressive symptoms (β = 0.025, −0.020–0.070; p = 0.281). Loneliness was strongly associated with subsequent increases in both depressive (β = 0.309, 0.159–0.459; p < 0.001) and anxiety (β = 0.301, 0.165–0.436; p < 0.001) symptoms. Compared to social distancing, adhering to stay-at-home orders or quarantining were not associated with anxiety and depressive symptoms or loneliness (both p ≥ 0.095).Conclusions: High loneliness may be a key risk factor for the development of future anxiety or depressive symptoms, underscoring the need to combat or prevent loneliness both throughout and beyond the COVID-19 pandemic. COVID-19 containment strategies were not associated with mental health, indicating that other factors may explain previous reports of mental health deterioration throughout the pandemic.

scholarly journals Capital adjustment cost and inconsistency in income-based dynamic panel models with fixed effects

2019 ◽  
Vol 20 (4) ◽  
pp. e1002-e1018 ◽  
Author(s):  
Parantap Basu ◽  
Yoseph Getachew ◽  
Keshab Bhattarai
Keyword(s):  
Fixed Effects ◽  
Cost Of Capital ◽  
Moving Average ◽  
Adjustment Cost ◽  
Dynamic Panel ◽  
Panel Models ◽  
Capital Adjustment ◽  
Long Run ◽  
Fixed Effects Estimator ◽  
Dynamic Panel Models

Abstract After the seminal work of Nickell (1981), a vast literature demonstrates the inconsistency of ‘conditional convergence’ estimator in income-based dynamic panel models with fixed effects when the time horizon (T) is short but the sample of countries (N) is large. Less attention is given to the economic root of inconsistency of the fixed effects estimator when T is also large. Using a variant of the Ramsey growth model with long-run adjustment cost of capital, we demonstrate that the fixed effects estimator of such models could be inconsistent when T is large. This inconsistency arises because of the long-run adjustment cost of capital which gives rise to a negative moving average coefficient in the error term. Income convergence will be thus overestimated. We theoretically characterize the order of this inconsistency. Our Monte Carlo simulation demonstrates that the size of the bias is substantial and it is greater in economies with higher capital adjustment costs. We show that the use of instrumental variables that take into account the presence of the negative moving average term in the error will overcome this bias.

BIAS REDUCTION FOR DYNAMIC NONLINEAR PANEL MODELS WITH FIXED EFFECTS

2011 ◽  
Vol 27 (6) ◽  
pp. 1152-1191 ◽  
Author(s):  
Jinyong Hahn ◽  
Guido Kuersteiner
Keyword(s):  
Monte Carlo ◽  
Nonlinear Dynamic ◽  
Fixed Effects ◽  
Bias Reduction ◽  
Small Sample ◽  
Dynamic Panel ◽  
Panel Models ◽  
Fixed Effects Estimator ◽  
Dynamic Panel Models

The fixed effects estimator of panel models can be severely biased because of well-known incidental parameter problems. It is shown that this bias can be reduced in nonlinear dynamic panel models. We consider asymptotics wherenandTgrow at the same rate as an approximation that facilitates comparison of bias properties. Under these asymptotics, the bias-corrected estimators we propose are centered at the truth, whereas fixed effects estimators are not. We discuss several examples and provide Monte Carlo evidence for the small sample performance of our procedure.

Transformed-likelihood estimators for dynamic panel models with a very small T

2020 ◽  
pp. 1-20
Author(s):  
Mark Pickup ◽  
Vincent Hopkins
Keyword(s):  
Panel Data ◽  
Random Effects ◽  
Fixed Effects ◽  
Dynamic Models ◽  
Dynamic Panel ◽  
Panel Models ◽  
Long Run ◽  
Individual Effects ◽  
Lagged Dependent Variable ◽  
Dynamic Panel Models

Conventional OLS fixed-effects and GLS random-effects estimators of dynamic models that control for individual-effects are known to be biased when applied to short panel data (T ≤ 10). GMM estimators are the most used alternative but are known to have drawbacks. Transformed-likelihood estimators are unused in political science. Of these, orthogonal reparameterization estimators are only tangentially referred to in any discipline. We introduce these estimators and test their performance, demonstrating that the unused orthogonal reparameterization estimator in particular performs very well and is an improvement on the commonly used GMM estimators. When T and/or N are small, it provides efficiency gains and overcomes the issues GMM estimators encounter in the estimation of long-run effects when the coefficient on the lagged dependent variable is close to one.

Sign in / Sign up

Export Citation Format

Share Document