About the Society
Papers, Posters, Syllabi
Submit an Item
Polmeth Mailing List
Below results based on the criteria 'Random Effects models'
Total number of records returned: 3
A Random Effects Approach to Legislative Ideal Point Estimation
random effects models
Conventionally, scholars use either standard probit/logit techniques or fixed-effect methods to estimate legislative ideal points. However, these methods are unsatisfactory when a limited number of votes are available: standard probit/logit methods are poorly equipped to handle multiple votes and fixed-effect models disregard serious ``incidental parameter'' problems. In this paper I present an alternative approach that moves beyond single-vote probit/logit analysis without requiring the large number of votes needed for fixed-effects models. The method is based on a random effects, panel logit framework that models ideal points as stochastic functions of legislator characteristics. Monte Carlo results and an application to trade politics demonstrate the practical usefulness of the method.
Explaining Fixed Effects: Random Effects modelling of Time-Series Cross-Sectional and Panel Data
Random Effects models
Fixed Effects models
Random coefficient models
Fixed effects vector decomposition
Time-Series Cross-Sectional Data
This article challenges Fixed Effects (FE) modelling as the ‘default’ for time-series-cross-sectional and panel data. Understanding differences between within- and between-effects is crucial when choosing modelling strategies. The downside of Random Effects (RE) modelling – correlated lower-level covariates and higher-level residuals – is omitted-variable bias, solvable with Mundlak’s (1978a) formulation. Consequently, RE can provide everything FE promises and more, and this is confirmed by Monte-Carlo simulations, which additionally show problems with another alternative, Plümper and Troeger’s Fixed Effects Vector Decomposition method, when data are unbalanced. As well as being able to model time-invariant variables, RE is readily extendable, with random coefficients, cross-level interactions, and complex variance functions. An empirical example shows that disregarding these extensions can produce misleading results. We argue not simply for technical solutions to endogeneity, but for the substantive importance of context and heterogeneity, modelled using RE. The implications extend beyond political science, to all multilevel datasets.
Political Regimes and Infant Death: Democratization and Its Consequences for Infant Mortality, 1970-2008.
Ramos, Antonio Pedro
random effects models
In this study, I investigate the causal linkage between political regimes and health outcomes over 180 countries between 1970 and 2010. While there are a number of previous empirical studies on this topic, the results of those studies are mixed and even contradictory. Missing data and measurement error present a major challenge. The main outcome of interest---child mortality--was until very recently poorly measured or unmeasured for many countries, specially for dictatorships and low-income countries. A lack of comparable measures of political regimes across time periods and countries also contributes to the contradictory findings in the existing literature. Finally, new statistical techniques that capture the important over-time dynamics that we expect to find in the translation of regime type into health outcomes have not previously been applied. Thus, though there are many examples of wealthy democracies with low infant mortality and high infant mortality countries are disproportionally autocratic, it remains unclear whether regime type causes lower levels of child mortality. Recently a group of health scholars compiled a high resolution data set based on 16,174 measurements of mortality rate of children younger than 5 years old for 187 countries from 1970 to 2009. There measurements are based upon information from all available sources, including vital registration systems, summary birth histories in censuses and surveys, and complete birth histories ([CITE]). I revisit the connection between regime type and infant mortality using this data set and flexible Bayesian statistical techniques that are specially tailored for the problem. To gain greater leverage on the causal effect of political regime on health outcomes, I focus on democratization episodes occurring since 1970. I present hierarchical longitudinal models tracking over-time changes in mean child mortality and investigate whether democratization episodes are followed by systematics changes from previous trend in infant mortality. I also apply matching techniques to compare changes in infant mortality following democratization episodes to change in infant mortality in similar countries which did not democratized during the same period. I find that there is little if any effect of democracy on health outcomes.