ppcheckPLMIX: Posterior predictive check for Bayesian mixtures of...
In PLMIX: Bayesian Analysis of Finite Mixtures of Plackett-Luce Models for Partial Rankings/Orderings

Description Usage Arguments Details Value Author(s) References See Also Examples

Perform posterior predictive check to assess the goodness-of-fit of Bayesian mixtures of Plackett-Luce models with a different number of components.

1 2	ppcheckPLMIX(pi_inv, seq_G, MCMCsampleP, MCMCsampleW, top1 = TRUE, paired = TRUE, parallel = FALSE)

`pi_inv`	An object of class `top_ordering`, collecting the numeric NxK data matrix of partial orderings, or an object that can be coerced with `as.top_ordering`.
`seq_G`	Numeric vector with the number of components of the Plackett-Luce mixtures to be assessed.
`MCMCsampleP`	List of size `length(seq_G)`, whose generic element is a numeric Lx(G*K) matrix with the MCMC samples of the component-specific support parameters.
`MCMCsampleW`	List of size `length(seq_G)`, whose generic element is a numeric LxG matrix with the MCMC samples of the mixture weights.
`top1`	Logical: whether the posterior predictive p-value based on the top item frequencies has to be computed. Default is `TRUE`.
`paired`	Logical: whether the posterior predictive p-value based on the paired comparison frequencies has to be computed. Default is `TRUE`.
`parallel`	Logical: whether parallelization should be used. Default is `FALSE`.

The ppcheckPLMIX function returns two posterior predictive p-values based on two chi squared discrepancy variables involving: (i) the top item frequencies and (ii) the paired comparison frequencies. In the presence of partial sequences in the pi_inv matrix, the same missingness patterns observed in the dataset (i.e., the number of items ranked by each sample unit) are reproduced on the replicated datasets from the posterior predictive distribution.

A list with a named element:

post_pred_pvalue

Numeric length(seq_G)x2 matrix of posterior predictive p-values based on the top item and paired comparison frequencies. If either top1 or paired argument is FALSE, the corresponding matrix entries are NA.

Cristina Mollica and Luca Tardella

Mollica, C. and Tardella, L. (2017). Bayesian Plackett-Luce mixture models for partially ranked data. Psychometrika, 82(2), pages 442–458, ISSN: 0033-3123, DOI: 10.1007/s11336-016-9530-0.

ppcheckPLMIX_cond

data(d_carconf)
K <- ncol(d_carconf)

## Fit 1- and 2-component PL mixtures via MAP estimation
MAP_1 <- mapPLMIX_multistart(pi_inv=d_carconf, K=K, G=1, 
                             n_start=2, n_iter=400*1)

MAP_2 <- mapPLMIX_multistart(pi_inv=d_carconf, K=K, G=2, 
                             n_start=2, n_iter=400*2)
                                   
MAP_3 <- mapPLMIX_multistart(pi_inv=d_carconf, K=K, G=3, 
                             n_start=2, n_iter=400*3)
                                   
mcmc_iter <- 30
burnin <- 10

## Fit 1- and 2-component PL mixtures via Gibbs sampling procedure
GIBBS_1 <- gibbsPLMIX(pi_inv=d_carconf, K=K, G=1, n_iter=mcmc_iter, 
                      n_burn=burnin, init=list(p=MAP_1$mod$P_map,
                      z=binary_group_ind(MAP_1$mod$class_map,G=1)))
GIBBS_2 <- gibbsPLMIX(pi_inv=d_carconf, K=K, G=2, n_iter=mcmc_iter, 
                      n_burn=burnin, init=list(p=MAP_2$mod$P_map,
                      z=binary_group_ind(MAP_2$mod$class_map,G=2)))
GIBBS_3 <- gibbsPLMIX(pi_inv=d_carconf, K=K, G=3, n_iter=mcmc_iter, 
                      n_burn=burnin, init=list(p=MAP_3$mod$P_map,
                      z=binary_group_ind(MAP_3$mod$class_map,G=3)))
                            
## Checking goodness-of-fit of the estimated mixtures
CHECK <- ppcheckPLMIX(pi_inv=d_carconf, seq_G=1:3, 
                      MCMCsampleP=list(GIBBS_1$P, GIBBS_2$P, GIBBS_3$P), 
                      MCMCsampleW=list(GIBBS_1$W, GIBBS_2$W, GIBBS_3$W))
CHECK$post_pred_pvalue