A Dirichlet process functional approach to heteroscedastic-consistent covariance estimation
From MaRDI portal
Publication:324685
DOI10.1016/J.IJAR.2016.07.008zbMATH OpenNonearXiv1602.05155OpenAlexW2284530505MaRDI QIDQ324685FDOQ324685
Authors: George Karabatsos 
Publication date: 17 October 2016
Published in: International Journal of Approximate Reasoning (Search for Journal in Brave)
Abstract: The mixture of Dirichlet process (MDP) defines a flexible prior distribution on the space of probability measures. This study shows that ordinary least-squares (OLS) estimator, as a functional of the MDP posterior distribution, has posterior mean given by weighted least-squares (WLS), and has posterior covariance matrix given by the (weighted) heteroscedastic-consistent sandwich estimator. This is according to a pairs bootstrap distribution approximation of the posterior, using a P'olya urn scheme. Also, when the MDP prior baseline distribution is specified as a product of independent probability measures, this WLS solution provides a new type of generalized ridge regression estimator which can handle multicollinear or singular design matrices even when the number of covariates exceeds the sample size, and which shrinks the coefficient estimates of irrelevant covariates towards zero, thus useful for nonlinear regressions. Also, this MDP/OLS functional methodology can be extended to methods for analyzing the sensitivity of the heteroscedasticity-consistent causal effect size over a range of hidden biases due to missing covariates omitted from the regression, and more generally extended to a Vibration of Effects analysis. The methodology is illustrated through the analysis of simulated and real data sets. Overall, this study establishes new connections between Dirichlet process functional inference, the bootstrap, consistent sandwich covariance estimation, ridge shrinkage regression, WLS, and sensitivity analysis, to provide regression methodology useful for inferences of the mean dependent response.
Full work available at URL: https://arxiv.org/abs/1602.05155
Cites Work
- Least angle regression. (With discussion)
- Bootstrap methods: another look at the jackknife
- Kernel stick-breaking processes
- An ANOVA Model for Dependent Random Measures
- Ridge Regression: Biased Estimation for Nonorthogonal Problems
- Title not available (Why is that?)
- Observational studies.
- A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity
- Mixtures of Dirichlet processes with applications to Bayesian nonparametric problems
- A Bayesian analysis of some nonparametric problems
- Improved heteroscedasticity-consistent covariance matrix estimators
- Asymptotic properties of a robust variance matrix estimator for panel data when \(T\) is large
- The Bias of a Heteroskedasticity Consistent Covariance Matrix Estimator
- Asymptotic Normality and Consistency of the Least Squares Estimators for Families of Linear Regressions
- Some priors for sparse regression modelling
- Microeconometrics
- Tuning Parameter Selection in High Dimensional Penalized Likelihood
- Title not available (Why is that?)
- Ferguson distributions via Polya urn schemes
- Theory and numerical analysis for exact distributions of functionals of a Dirichlet process
- Bayesian regression with nonparametric heteroskedasticity
- Risk of Bayesian inference in misspecified models, and the sandwich covariance matrix
- Bayesian Interpretations of Heteroskedastic Consistent Covariance Estimators Using the Informed Bayesian Bootstrap
- Conjugacy as a Distinctive Feature of the Dirichlet Process
- Bayesian sandwich posteriors for pseudo-true parameters
- Considerazioni generali sull’impostazione bayesiana di problemi non parametrici
- Parametric fractional imputation for nonignorable missing data
- Nonparametric Bayesian analysis of a proportion for a small area under nonignorable nonresponse
- Beta-product dependent Pitman-Yor processes for Bayesian inference
- A nonparametric Bayesian prediction interval for a finite population mean
- A multivariate extension of a vector of two-parameter Poisson–Dirichlet processes
- Model-robust regression and a Bayesian ``sandwich estimator
- Vectors of two-parameter Poisson-Dirichlet processes
- Distributional properties of means of random probability measures
Uses Software
This page was built for publication: A Dirichlet process functional approach to heteroscedastic-consistent covariance estimation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q324685)
