About the Society
Papers, Posters, Syllabi
Submit an Item
Polmeth Mailing List
Below results based on the criteria 'Endogeneity'
Total number of records returned: 5
Advancement in the House of Representatives
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.
Two-Stage Estimation of Non-Recursive Choice Models
Alvarez, R. Michael
two-stage probit least squares
two-stage conditional maximum likelihood
Questions of causation are important issues in empirical research on political behavior. Most of the discussion of the econometric problems associated with multi-equation models with reciprocal causation has focused on models with continuous dependent variables (e.g. Markus and Converse 1979; Page and Jones 1979). Yet many models of political behavior involve discrete or dichotomous dependent variables; this paper describes two techniques which can consistently estimate reciprocal relationships between dichotomous and continuous dependent variables. The first, two-stage probit least squares (2SPLS), is very similar to two-stage instrumental variable techniques. The second, two-stage conditional maximum likelihood (2SCML), may overcome problems associated with 2SPLS, but has not been used in the political science literature. First, we demonstrate the potential pitfalls of ignoring the problems of reciprocal causation in non-recursive choice models. Then, we show the properties of both techniques using Monte Carlo simulations: both the two-stage models perform well in large samples, but in small samples the 2SPLS model has superior statistical properties. However, the 2SCML model offers an explicit statistical test for endogeneity. Last, we apply these techniques to an empirical example which focuses on the relationship between voter preferences in a presidential election and the voter's uncertainty about the policy positions taken by the candidates. This example demonstrates the importance of these techniques for political science research.
The Revolution Against Affirmative Action in California: Politics, Economics, and Proposition 209
Alvarez, R. Michael
Butterfield, Tara L.
generalized extreme value
race and politics
We consider two possible explanations --- economic anxiety and racial division --- for the appeal of Proposition 209 to California voters during the 1996 election. To test these hypotheses, we analyze voter exit poll data from teh 1996 California election. We utiliize a two--stage logit model to allow for the endogeneity of candidate endorsements. We find support for the second of our two hypotheses, which leads us to conclude that racial division fueled by a fear of arbitrary exclusion prompted voter support for Proposition 209.
Boundary that Matters? Causal Inference of the School Quality Effect on Land Prices
omitted variable bias
According to the hedonic model, the effect of areal policy such as school quality is reflected, or capitalized, in land prices. The conventional OLS, however, suffers from endogeneity bias, measurement error, and omitted variable bias. To solve these problems, this paper proposes spatial differences-in-differences (DID). We match literally the nearest two sample points in a small block to make a pair. If the two points belong to different school-attendance areas, the pair is a treatment pair. Otherwise, the pair is a control pair. If school quality matters for land price, the variance of pairwise land price gap of the treatment pairs should be larger than that of the control pairs. Another new method, spatial and temporal DID, exploits introduction of school choice program to improve robustness against omitted variable bias. When applying these methods to data of Tokyo, F-test fails to reject the null hypothesis. We show, however, that land use zoning and floor area ratio have effects on land price by using spatial DID.