Skip to contents

Overview

The muppet package executes Measurement and Uncertainty Preserving ParamETric (MUPPET) modeling. MUPPET is a modular approach to model fitting, using Bayesian approaches to modeling and estimation. Models and analyses are conceived of in terms of fragments. Each may be fit to data or analyzed, possibly conditional on antecedent fragments. In that case, the results from the antecedent fragments are carried forward to those that depend on them.

Installation

You can install the current public version of muppet from GitHub. It depends on other packages for certain fuctionality. muppet uses functions that are part of the mcmcplots package, developed by S. McKay Curtis, to plot results from the chains. However, mcmcplots is no longer availabe on CRAN. A verion of mcmcplots may be installed from GitHub, and then muppet can also be installed from GitHub. This may be done as follows:

# install.packages("remotes")  # if you don't have remotes installed

# Install mcmcplots from maintained fork
remotes::install_github("Roy-Levy/mcmcplots")

# Then install muppet
remotes::install_github("Roy-Levy/muppet_pub")

Examples

The Examples and Tips page has some examples and explanations of some of the basic functionality.

Contact

Have questions or suggestions for features? Found a bug? Please contact me at