fitlmer.mcmc: Run mixed-effects model ('lmer') with MCMC p-value
In dalejbarr/simgen: Monte Carlo Simulations of Simultaneous Generalization

Description Usage Arguments Details Note See Also

Uses lmer to fit a random-intercepts only model, and then uses mcmcsamp to obtain p-values (one-factor design only).

1	fitlmer.mcmc(mcr.data, wsbi, nmcmc = 10000)

`mcr.data`	A dataframe formatted as described in `mkDf`.
`wsbi`	Whether the design is between-items (TRUE) or within-items (FALSE); has no effect because the model is random-intercepts only, but was included for consistency with `fitlmer` and `fitanova`.
`nmcmc`	Number of Markov-Chain Monte Carlo simulations (default = 10000).

Tries to fit the most complex model requested, and if that fails to converge, then tries progressively simpler models, before calculating mcmc p-value. If no model converges, returns NAs. The MCMC procedure is based on Baayen's pvals.fnc in package languageR.

This function no longer works and will throw an error, as the mcmcsamp function was removed starting with lme4 package 1.0; see mcmcsamp for details.

fitlmer

dalejbarr/simgen documentation built on May 14, 2019, 3:32 p.m.