About the Society
Papers, Posters, Syllabi
Submit an Item
Polmeth Mailing List
Below results based on the criteria 'nonparametric'
Total number of records returned: 8
Nonparametric Priors For Ordinal Bayesian Social Science Models: Specification and Estimation
generalized linear mixed model
Dirichlet process mixture models
nonparametric Bayesian inference
A generalized linear mixed model, ordered probit, is used to estimate levels of stress in presidential political appointees as a means of understanding their surprisingly short tenures. A Bayesian approach is developed, where the random effects are modeled with a Dirichlet process mixture prior, allowing for useful incorporation of prior information, but retaining some vagueness in the form of the prior. Applications of Bayesian models in the social sciences are typically done with ``noninformative'' priors, although some use of informed versions exists. There has been disagreement over this, and our approach may be a step in the direction of satisfying both camps. We give a detailed description of the data, show how to implement the model, and describe some interesting conclusions. The model utilizing a nonparametric prior fits better and reveals more information in the data than standard approaches.
Causal Inference with Differential Measurement Error: Nonparametric Identification and Sensitivity Analyses of a Field Experiment on Democratic Deliberations
Political scientists have long been concerned about the validity of survey measurements. Although many have studied classical measurement error in linear regression models where the error is assumed to arise completely at random, in a number of situations the error may be correlated with the outcome. We analyze the impact of differential measurement error on causal estimation. The proposed nonparametric identification analysis avoids arbitrary modeling decisions and formally characterizes the roles of additional assumptions. We show the serious consequences of differential misclassification and offer a new sensitivity analysis that allows researchers to evaluate the robustness of their conclusions. Our methods are motivated by a field experiment on democratic deliberations, in which one set of estimates potentially suffers from differential misclassification. We show that an analysis ignoring differential measurement error may considerably overestimate the causal effects. This finding contrasts with the case of classical measurement error which always yields attenuation bias.
An Alternative Solution to the Heckman Selection Problem: Selection Bias as Functional Form Misspecification
functional form misspecification
The "selection problem" is typically seen as a form of omitted variable bias. We recast the problem as one of functional form misspecification and examine two situations in which flexible or nonparametric estimation techniques may be used as a complement or alternative to traditional selection models. First, we show that such techniques can allow a researcher to recover the conditional relationship between covariates and the expected outcome, even if data on the probability of selection into the subsample is unavailable. We demonstrate the validity of this approach analytically and using Monte Carlo simulations. Second, we show that flexible methods can be used to validate or improve a linear selection model specification when a researcher does possess the prior-stage data. We illustrate this process with an application to data from Mroz (1987) on women's wages.
Definition and Diagnosis of Problematic Attrition in Randomized Controlled Experiments
Martel García, Fernando
randomized controlled experiments
directed acyclic graphs
average treatment effect
Attrition is the Achilles' Heel of the randomized experiment: It is fairly common, and it can completely unravel the benefits of randomization. Using the structural language of causal diagrams I demonstrate that attrition is problematic for identification of the average treatment effect (ATE) if -- and only if -- it is a common effect of the treatment and the outcome (or a cause of the outcome other than the treatment). I also demonstrate that whether the ATE is identified and estimable for the full population of units in the experiment, or only for those units with observed outcomes, depends on two d-separation conditions. One of these is testable ex-post under standard experimental assumptions. The other is testable ex-ante so long as adequate measurement protocols are adopted. Missing at Random (MAR) assumptions are neither necessary nor sufficient for identification of the ATE.
Discriminating Methods: Tests for Nonnested Discrete Choice Models
Clarke, Kevin A.
Signorino, Curtis S.
We consider the problem of choosing between rival models that are nonnested in terms of their functional forms. We discuss both a parametric and distribution-free procedure for making this choice, and demonstrate through a monte carlo simulation that discrimination is possible. The results of the simulation also allow us to compare the relative power of the two tests.
Rational Expectations Coordinating Voting in American Presidential and House Elections
Mebane, Walter R.
generalized extreme value model
Monte Carlo integration
I define a probabilistic model of individuals' presidential-year vote choices for President and for the House of Representatives in which there is a coordinating (Bayesian Nash) equilibrium among voters based on rational expectations each voter has about the election outcomes. I estimate the model using data from the six American National Election Study Pre-/Post-Election Surveys of years 1976--1996. The coordinating model passes a variety of tests, including a test against a majoritarian model in which there is rational ticket splitting but no coordination. The results give strong individual-level support to Alesina and Rosenthal's theory that voters balance institutions in order to moderate policy. The estimates describe vote choices that strongly emphasize the presidential candidates. I also find that a voter who says economic conditions have improved puts more weight on a discrepancy between the voter's ideal point and government policy with a Democratic President than on a discrepancy of the same size with a Republican President.
Bayesian and Likelihood Inference for 2 x 2 Ecological Tables: An Incomplete Data Approach
Missing information principle
Nonparametric Bayesian Modeling.
Ecological inference is a statistical problem where aggregate-level data are used to make inferences about individual-level behavior. Recent years have witnessed resurgent interest in ecological inference among political methodologists and statisticians. In this paper, we conduct a theoretical and empirical study of Bayesian and likelihood inference for 2 x 2 ecological tables by applying the general statistical framework of incomplete data. We first show that the ecological inference problem can be decomposed into three factors: distributional effects which address the possible misspecification of parametric modeling assumptions about the unknown distribution of missing data, contextual effects which represent the possible correlation between missing data and observed variables, and aggregation effects which are directly related to the loss of information caused by data aggregation. We then examine how these three factors affect inference and offer new statistical methods to address each of them. To deal with distributional effects, we propose a nonparametric Bayesian model based on a Dirichlet process prior which relaxes common parametric assumptions. We also specify the statistical adjustments necessary to account for contextual effects. Finally, while little can be done to cope with aggregation effects, we offer a method to quantify the magnitude of such effects in order to formally assess its severity. We use simulated and real data sets to empirically investigate the consequences of these three factors and to evaluate the performance of our proposed methods. C code, along with an easy-to-use R interface, is publicly available for implementing our proposed methods.
A New Non-Parametric Matching Method for Bias Adjustment with Applications to Economic Evaluations
semiparametric and nonparametric matching methods
randomized controlled trials
health economic evaluation
In health economic studies that use observational data, a key concern is how to adjust for imbalances in baseline covariates due to the non-random assignment of the programs under evaluation. Traditional methods of covariate adjustment such as regression and propensity score matching are model dependent and often fail to replicate the results of randomized controlled trials. We demonstrate a new non-parametric matching method, Genetic Matching, which is a generalization of propensity score and Mahalanobis distance matching, using two contrasting case studies. In the first, an economic evaluation of a clinical intervention (Pulmonary Artery Catheterization), applying Genetic Matching to observational data replicates the substantive results of a corresponding randomized controlled trial unlike the extant literature. And in the second case study evaluating capitation versus fee-for service, Genetic Matching radically improves balance on baseline covariates and overturns previous conclusions based on traditional methods.