pmwg-package: pmwg: Particle Metropolis Within Gibbs.

In pmwg: Particle Metropolis Within Gibbs

pmwg: Particle Metropolis Within Gibbs.

Description

The pmwg package provides a general purpose implementation of the sampling techniques outlined in Gunawan et al. (2020) \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1016/j.jmp.2020.102368")}. The user of this package is required to provide their own log likelihood function, but given this the functions provided can estimate model parameters, the full covariance matrix and subject random effects in a hierarchical Bayesian way.

Documentation

The documentation found at contains background information and motivation for the approach used in this package and several detailed examples of the package in action. It also includes a list of common problems and associated troubleshooting steps.

User input

The user is expected to provide a data source in a format that is compatible with R data.frame methods. This data must have at least one column named ‘subject' that has a unique identifier for each subject’s data.

Additionally the user should provide a function that when given a set of parameter estimates and the data for a single subject return the log of the likelihood of that data given the parameter estimates.

The final piece of required information is a list of the names of each parameter that should be estimated. There is also the capability to provide priors on the model parameters, start points for the model parameters and covariance matrix as well as options to fine tune the sampling process

Author(s)

Maintainer: Gavin Cooper [e-mail] (Package creator and maintainer) [translator]

Authors:

• Reilly Innes [e-mail]

• Caroline Kuhne [e-mail]

• Jon-Paul Cavallaro [e-mail]

• David Gunawan [e-mail] (Author of original MATLAB code)

• Guy Hawkins [e-mail]

• Scott Brown [e-mail] (Original translation from MATLAB to R) [translator]

• Niek Stevenson [e-mail]

References

Gunawan, D., Hawkins, G. E., Tran, M. N., Kohn, R., & Brown, S. D. (2020). New estimation approaches for the hierarchical Linear Ballistic Accumulator model. Journal of Mathematical Psychology, 96, 102368.

See Also

Useful links:

• https://github.com/university-of-newcastle-research/pmwg

• Report bugs at https://github.com/university-of-newcastle-research/pmwg/issues

pmwg documentation built on May 29, 2024, 12:09 p.m.