STAT 544 Applets

Please run these apps from r using the runGitHub function from the shiny package:

install.packages('shiny')

One parameter conjugate analysis

This app allows fully Bayesian inference using conjugate priors for univariate models including binomial, Poisson, normal, and exponential. To run this app use

shiny::runGitHub('jarad/one_parameter_conjugate')

Mean height (prior elicitation activity)

This app is used in STAT 544 to illustrate the process of eliciting a prior and then also to demonstrate the difference between confidence intervals and credible intervals. If you are currently enrolled in STAT 544, you should not access this app until after we use it in class (just to avoid anchoring). To run this app use

shiny::runGitHub('jarad/prior_elicitation_activity/')

Pvalue interpretation

This app illustrates the difficulty in interpreting pvalues by simulating data sets within the context of a normal model with unknown variance and testing the mean to be zero. The user selects a pvalue (technically a pvalue range) and the software will calculate what proportion of time the null hypotheses is true over a set of simulations with that pvalue. To run this app use

shiny::runGitHub('jarad/pvalue')