Time-invariant regressors under fixed effects: Simple identification via a proxy variable

2020 ◽  
Vol 186 ◽  
pp. 108799
Author(s):  
Matěj Bělín
Keyword(s):  
Fixed Effects ◽  
Proxy Variable ◽  
Time Invariant

Dental Care Use, Edentulism, and Systemic Health among Older Adults

2021 ◽  
pp. 002203452110190
Author(s):  
C.D. Meyerhoefer ◽  
J.V. Pepper ◽  
R.J. Manski ◽  
J.F. Moeller
Keyword(s):  
Older Adults ◽  
Heart Disease ◽  
Chronic Illness ◽  
Dental Care ◽  
Fixed Effects ◽  
Past Research ◽  
Positive Correlations ◽  
Systemic Health ◽  
Time Invariant ◽  
Retirement Study

Past research suggests there are systematic associations between oral health and chronic illness among older adults. Although causality has not yet been credibly established, periodontitis has been found to be associated with higher risk of both heart disease and stroke. We advance this literature by estimating the direct association between dental care use and systemic health using multiple waves of the 1992 to 2016 Health and Retirement Study. Through the inclusion of individual fixed effects in our regression models, we account for unobservable time-invariant characteristics of individuals that might otherwise bias estimates of the association between dental care use and health. We find statistically significant negative associations between dental care use and the number of health conditions, self-reported overall health, the incidence of heart disease, and the incidence of stroke. In particular, the use of dental care within the past 2 y is associated with a 2.7% reduction in the likelihood of being diagnosed with a heart condition and a reduction in the likelihood of a stroke diagnosis of between 5.3% and 11.6%. We also find large positive correlations between edentulism and the measures of chronic illness. Associations from models estimated separately for men and women are qualitatively similar to one another. These findings provide additional motivation for the consideration of a Medicare dental benefit.

scholarly journals Fixed Effects Vector Decomposition: A Magical Solution to the Problem of Time-Invariant Variables in Fixed Effects Models?

2011 ◽  
Vol 19 (2) ◽  
pp. 135-146 ◽  
Author(s):  
William Greene
Keyword(s):  
Fixed Effects ◽  
Ordinary Least Squares ◽  
Fe Model ◽  
Real World Data ◽  
Problem Of Time ◽  
Dummy Variable ◽  
Vector Decomposition ◽  
Step Procedure ◽  
Time Invariant ◽  
Subsequent Literature

Plümper and Troeger (2007) propose a three-step procedure for the estimation of a fixed effects (FE) model that, it is claimed, “provides the most reliable estimates under a wide variety of specifications common to real world data.” Their fixed effects vector decomposition (FEVD) estimator is startlingly simple, involving three simple steps, each requiring nothing more than ordinary least squares (OLS). Large gains in efficiency are claimed for cases of time-invariant and slowly time-varying regressors. A subsequent literature has compared the estimator to other estimators of FE models, including the estimator of Hausman and Taylor (1981) also (apparently) with impressive gains in efficiency. The article also claims to provide an efficient estimator for parameters on time-invariant variables (TIVs) in the FE model. None of the claims are correct. The FEVD estimator simply reproduces (identically) the linear FE (dummy variable) estimator then substitutes an inappropriate covariance matrix for the correct one. The consistency result follows from the fact that OLS in the FE model is consistent. The “efficiency” gains are illusory. The claim that the estimator provides an estimator for the coefficients on TIVs in an FE model is also incorrect. That part of the parameter vector remains unidentified. The “estimator” relies upon a strong assumption that turns the FE model into a type of random effects model.

Forced Displacement in Colombia: What Determines Expulsion and Reception in Municipalities?

2014 ◽  
Vol 20 (4) ◽  
pp. 585-597 ◽  
Author(s):  
Ximena Dueñas ◽  
Paola Palacios ◽  
Blanca Zuluaga
Keyword(s):  
Panel Data ◽  
Fixed Effects ◽  
Independent Variables ◽  
Panel Data Estimation ◽  
Vector Decomposition ◽  
Displaced People ◽  
Time Invariant ◽  
Data Estimation ◽  
Conflict Intensity ◽  
Municipal Level

AbstractThis document explores the expulsion and reception determinants of displaced people among Colombian municipalities. For this purpose, we use fixed effects panel data estimations for the period 2004–2009, with municipality year as the unit of analysis. To the best of our knowledge, this is the first paper in Colombia that focuses on reception and the first one using panel data at municipal level to explain expulsion and reception. We find that, contrary to what one may expect, some independent variables affect both expulsion and reception of displaced people in the same direction; for instance, municipalities where homicide rates and conflict intensity are high, are associated with both higher reception and expulsion rates. In addition to the conventional panel data estimation, we also run a fixed effect vector decomposition to identify the explicit effects of certain time-invariant variables.

Spatial Errors in Count Data Regressions

2016 ◽  
Vol 5 (1) ◽  
Author(s):  
Marinho Bertanha ◽  
Petra Moser
Keyword(s):  
Cross Section ◽  
Count Data ◽  
Spatial Dependence ◽  
Fixed Effects ◽  
Negative Binomial ◽  
Variance Estimator ◽  
Research Fields ◽  
Patent Counts ◽  
The Cross ◽  
Time Invariant

AbstractCount data regressions are an important tool for empirical analyses ranging from analyses of patent counts to measures of health and unemployment. Along with negative binomial, Poisson panel regressions are a preferred method of analysis because the Poisson conditional fixed effects maximum likelihood estimator (PCFE) and its sandwich variance estimator are consistent even if the data are not Poisson-distributed, or if the data are correlated over time. Analyses of counts may however also be affected by correlation in the cross-section. For example, patent counts or publications may increase across related research fields in response to common shocks. This paper shows that the PCFE and its sandwich variance estimator are consistent in the presence of such dependence in the cross-section – as long as spatial dependence is time-invariant. We develop a test for time-invariant spatial dependence and provide code in STATA and MATLAB to implement the test.

scholarly journals The Impact of Job Stress on Smoking and Quitting: Evidence from the HRS

2010 ◽  
Vol 10 (1) ◽  
Author(s):  
Padmaja Ayyagari ◽  
Jody L Sindelar
Keyword(s):  
Job Stress ◽  
Fixed Effects ◽  
Smoking Behavior ◽  
Occupational Characteristics ◽  
Secular Decline ◽  
Related Stress ◽  
Job Related Stress ◽  
Time Invariant ◽  
The Impact ◽  
Invariant Factors

Abstract Job-related stress might affect smoking behavior because smoking may relieve stress and stress can make individuals more present-focused. Alternatively, individuals may both self-select into stressful jobs and choose to smoke based on unobserved factors. We use data from the Health and Retirement Study to examine how job stress affects the probability that smokers quit and the number of cigarettes smoked for current smokers. To address the potential endogeneity of job stress based on time invariant factors, we include individual fixed effects, which control for factors such as ability to handle stress. Occupational fixed effects are also included to control for occupational characteristics other than stress; time dummies control for the secular decline in smoking rates. Using a sample of people who smoked in the previous wave, we find that job stress is positively related to continuing to smoke among recent smokers. The results indicate that the key impact of stress is on the extensive margin of smoking, as opposed to the number of cigarettes smoked.

scholarly journals feologit: A new command for fitting fixed-effects ordered logit models

2020 ◽  
Vol 20 (2) ◽  
pp. 253-275 ◽  
Author(s):  
Gregori Baetschmann ◽  
Alexander Ballantyne ◽  
Kevin E. Staub ◽  
Rainer Winkelmann
Keyword(s):  
Fixed Effects ◽  
Generalized Method Of Moments ◽  
Logit Models ◽  
Ordered Logit ◽  
The Social ◽  
The Standard Model ◽  
Dependent Variables ◽  
Using Data ◽  
New Community ◽  
Time Invariant

In this article, we describe how to fit panel-data ordered logit models with fixed effects using the new community-contributed command feologit. Fixed-effects models are increasingly popular for estimating causal effects in the social sciences because they flexibly control for unobserved time-invariant heterogeneity. The ordered logit model is the standard model for ordered dependent variables, and this command is the first in Stata specifically for this model with fixed effects. The command includes a choice between two estimators, the blowup and cluster (BUC) estimator introduced in Baetschmann, Staub, and Winkelmann (2015, Journal of the Royal Statistical Society, Series A 178: 685–703) and the BUC- τ estimator in Baetschmann (2012, Economics Letters 115: 416–418). Baetschmann, Staub, and Winkelmann (2015) showed that the BUC estimator has good properties and is almost as efficient as more complex estimators such as generalized method-of-moments and empirical likelihood estimators. The command and model interpretations are illustrated with an analysis of the effect of parenthood on life satisfaction using data from the German Socio-Economic Panel.

scholarly journals Two-Way, One-Way or Dead-End Streets? Financial and Social Causes and Consequences of Generalized Trust

2021 ◽  
Author(s):  
Stefan Leenheer ◽  
Maurice Gesthuizen ◽  
Michael Savelkoul
Keyword(s):  
Fixed Effects ◽  
Life Experiences ◽  
Unobserved Heterogeneity ◽  
Panel Study ◽  
Adult Life ◽  
Causal Relationships ◽  
Generalized Trust ◽  
Social Contacts ◽  
Age Range ◽  
Time Invariant

AbstractScholars disagree on whether and to what extent adult life experiences can influence generalized trust and vice versa. Going beyond the methodological limitations of former studies, we aimed to answer the question as to what extent reciprocal causal relationships exist between generalized trust and the adult life experiences of financial success and (in)formal social contacts. We used two-wave cross-lagged panel models to identify those reciprocal causal relationships, and fixed-effects models to assess if they might be biased due to unaccounted time-invariant influences. Data from the Dutch NELLS panel study (age range 17–49) show that compelling empirical evidence is found for a reciprocal causal relationship between generalized trust and household income that does not suffer from bias due to unobserved heterogeneity. Furthermore, more trusting individuals experience a stronger decrease in material deprivation, but not vice versa. Trust and (in)formal social contacts are not causally related in any of our models.

scholarly journals Not so Harmless After All: The Fixed-Effects Model

2018 ◽  
Vol 27 (1) ◽  
pp. 21-45 ◽  
Author(s):  
Thomas Plümper ◽  
Vera E. Troeger
Keyword(s):  
Model Selection ◽  
Fixed Effects ◽  
Time Varying ◽  
Multiple Model ◽  
Fixed Effects Model ◽  
Omitted Variables ◽  
Hausman Test ◽  
Fixed Effects Estimator ◽  
Time Invariant

The fixed-effects estimator is biased in the presence of dynamic misspecification and omitted within variation correlated with one of the regressors. We argue and demonstrate that fixed-effects estimates can amplify the bias from dynamic misspecification and that with omitted time-invariant variables and dynamic misspecifications, the fixed-effects estimator can be more biased than the ‘naïve’ OLS model. We also demonstrate that the Hausman test does not reliably identify the least biased estimator when time-invariant and time-varying omitted variables or dynamic misspecifications exist. Accordingly, empirical researchers are ill-advised to rely on the Hausman test for model selection or use the fixed-effects model as default unless they can convincingly justify the assumption of correctly specified dynamics. Our findings caution applied researchers to not overlook the potential drawbacks of relying on the fixed-effects estimator as a default. The results presented here also call upon methodologists to study the properties of estimators in the presence of multiple model misspecifications. Our results suggest that scholars ought to devote much more attention to modeling dynamics appropriately instead of relying on a default solution before they control for potentially omitted variables with constant effects using a fixed-effects specification.

Sign in / Sign up

Export Citation Format

Share Document