Discussion: Posterior Inference in Bayesian Quantile Regression with Asymmetric Laplace Likelihood
نویسندگان
چکیده
To bake a Bayesian π (posterior) I was taught that you needed an L (likelihood) and a p (prior) – Oh yeah, and probably some data, don’t forget the data! So it comes as something of a shock to discover that there are 5,240 web documents employing the phrase “Bayesian quantile regression,” as of September 1, 2015, according to Google. Quantile regression would seem to be the very antithesis of a likelihood based procedure, committing the investigator to a parametric model for one paltry conditional quantile function, while professing total ignorance, even indifference, about the rest of the Deus ex machina, aka data generating mechanism.
منابع مشابه
Posterior Inference in Bayesian Quantile Regression with Asymmetric Laplace Likelihood
The paper discusses the asymptotic validity of posterior inference of pseudo-Bayesian quantile regression methods with complete or censored data when an asymmetric Laplace likelihood is used. The asymmetric Laplace likelihood has a special place in the Bayesian quantile regression framework because the usual quantile regression estimator can be derived as the maximum likelihood estimator under ...
متن کاملDiscussion: Posterior inference in Bayesian quantile regression with asymmetric Laplace likelihood
We begin by congratulating Professors Yang, Wang & He (hereafter YW&H) for a top-notch contribution to the literature, and thanking Professor Hallin for giving us the opportunity to discuss their work. We found it quite satisfying to discover that the MCMC algorithms over which we have toiled can be used to make valid asymptotic inference! In this discussion, we focus on MCMC algorithms for exp...
متن کاملA Bayesian Quantile Regression Analysis of Potential Risk Factors for Violent Crimes in USA
Bayesian quantile regression has drawn more attention in widespread applications recently. Yu and Moyeed (2001) proposed an asymmetric Laplace distribution to provide likelihood based mechanism for Bayesian inference of quantile regression models. In this work, the primary objective is to evaluate the performance of Bayesian quantile regression compared with simple regression and quantile regre...
متن کاملA New Family of Error Distributions for Bayesian Quantile Regression
We propose a new family of error distributions for model-based quantile regression, which is constructed through a structured mixture of normal distributions. The construction enables fixing specific percentiles of the distribution while, at the same time, allowing for varying mode, skewness and tail behavior. It thus overcomes the severe limitation of the asymmetric Laplace distribution – the ...
متن کاملBayesian semiparametric additive quantile regression
Quantile regression provides a convenient framework for analyzing the impact of covariates on the complete conditional distribution of a response variable instead of only the mean. While frequentist treatments of quantile regression are typically completely nonparametric, a Bayesian formulation relies on assuming the asymmetric Laplace distribution as auxiliary error distribution that yields po...
متن کامل