- Use *gpda_0.1.8.3*. It has the LBA-PDA, PLBA00, and PLBA1f models.
- Use *ggdmc_0.1.6.3*. It has special functions to call CUDA codes from R and send back to C++. It also has DE-MCMC *crossover* sampler and genetic algorithm, *migration*, sampler, in C++. I used it to fit Bayesian LBA model. It also fits Bayesian DDM model. The special functions in *ggdmc* calls CUDA C from R and sends the computation back to C++. This version shortens the PLBA computation time (31 participants) from almost half year (using more than 1 million model simulation) to around 1 month.
- *ggdmc_0.1.6.3* works with *gpda_0.1.8.3*. You also need a Nvidia CUDA compatible GPU.
- *dmc-one-plba1.tgz* zips DMC and an example, showing a prototype to recover parameters in PLBA1f model.
- I re-wrote `Dstats.dmc` and `IC.dmc` functions in DMC and put them in `util.R` to calculate specifically only DIC values. The two DMC functions can be found in *dmc-amsterdam17.zip*.
- Simulation data are oversized. Email yishinlin001@gmail.com, if you want to have these data.
- Please cite the paper and/or this OSF repository, if you use the codes here for your projects.
- To install cpda package, one must install two pre-requistic libraries, (1) GNU Scientific Library and (2) Boost. These libraries do not work on Windows platform without going through rather tedious tweaking.
- The ggdmc version of 0.2.3.2 would encounter the error: "Error in loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()), versionCheck = vI[[j]]) :". This means you need to install the package: ‘tmvtnorm’. It has been fixed in the newer version of ggdmc currently on CRAN.
**Reference**:
- R Core Team (2017). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
- Holmes, W., Trueblood, J. S., & Heathcote, A. (2016). A new framework for
modeling decisions about changing information: The Piecewise Linear Ballistic
Accumulator model. Cognitive Psychology, 85, 1--29,
doi: http://dx.doi.org/10.1016/j.cogpsych.2015.11.002.
