Chapter 6 Univariate regression

We describe how to perform Bayesian inference in some of the most common univariate models: normal-inverse gamma, logit, probit, multinomial probit and logit, ordered probit, negative binomial, tobit, quantile regression, and Bayesian bootstrap in linear models. The point of departure is assuming a random sample of cross-sectional units. We then show the posterior distributions of the parameters and some applications. In addition, we show how to perform inference in various models using three levels of programming skills: our graphical user interface (GUI), packages from R, and programming the posterior distributions. The first requires no programming skills, the second requires an intermediate level, and the third demands more advanced skills. We also include mathematical and computational exercises.

We can run our GUI typingshiny::runGitHub("besmarter/BSTApp", launch.browser=T) in the R console or any R code editor and execute it. However, users should see Chapter 5 for details.