DPpackage (version 1.1-6)

DPrasch: Bayesian analysis for a semiparametric Rasch model

Description

This function generates a posterior density sample for a semiparametric Rasch model, using a DP or a MDP prior for the distribution of the random effects.

Usage

DPrasch(y,prior,mcmc,offset,state,status, grid=seq(-10,10,length=1000),data=sys.frame(sys.parent()), compute.band=FALSE)

Arguments

y

a matrix giving the data for which the Rasch Model is to be fitted.

prior

a list giving the prior information. The list includes the following parameter: a0 and b0 giving the hyperparameters for prior distribution of the precision parameter of the Dirichlet process prior, alpha giving the value of the precision parameter (it must be specified if a0 is missing), mub and Sb giving the hyperparameters of the normal prior distribution for the mean of the normal baseline distribution, mu giving the mean of the normal baseline distribution (is must be specified if mub and Sb are missing), tau1 and tau2 giving the hyperparameters for the prior distribution of variance of the normal baseline distribution, sigma2 giving the variance of the normal baseline distribution (is must be specified if tau1 and tau2 are missing), and beta0 and Sbeta0 giving the hyperparameters of the normal prior distribution for the difficulty parameters.

mcmc

a list giving the MCMC parameters. The list must include the following integers: nburn giving the number of burn-in scans, nskip giving the thinning interval, nsave giving the total number of scans to be saved, and ndisplay giving the number of saved scans to be displayed on screen (the function reports on the screen when every ndisplay iterations have been carried out).

offset

this can be used to specify an a priori known component to be included in the linear predictor during the fitting.

state

a list giving the current value of the parameters. This list is used if the current analysis is the continuation of a previous analysis.

status

a logical variable indicating whether this run is new (TRUE) or the continuation of a previous analysis (FALSE). In the latter case the current value of the parameters must be specified in the object state.

grid

grid points where the density estimate is evaluated. The default is seq(-10,10,length=1000).

data

data frame.

compute.band

logical variable indicating whether the confidence band for the CDF must be computed.

Value

An object of class DPrasch representing the Rasch model fit. Generic functions such as print, plot, and summary have methods to show the results of the fit. The results include beta, mu, sigma2, the precision parameter alpha, and the number of clusters.The function DPrandom can be used to extract the posterior mean of the random effects.The list state in the output object contains the current value of the parameters necessary to restart the analysis. If you want to specify different starting values to run multiple chains set status=TRUE and create the list state based on this starting values. In this case the list state must include the following objects:

ncluster

an integer giving the number of clusters.

alpha

giving the value of the precision parameter.

b

a vector of dimension nsubjects giving the value of the random effects for each subject.

bclus

a vector of dimension nsubjects giving the value of the random effects for each clusters (only the first ncluster are considered to start the chain).

ss

an interger vector defining to which of the ncluster clusters each subject belongs.

beta

giving the value of the difficulty parameters.

mu

giving the mean of the normal baseline distributions.

sigma2

giving the variance of the normal baseline distributions.

Details

This generic function fits a semiparametric Rasch model as in San Martin et al. (2011), where the linear predictor is modeled as follows: $$\eta_{ij} = \theta_i - \beta_j, i=1,\ldots,n, j=1,\ldots,k$$ $$\theta_i | G \sim G$$ $$G | \alpha, G_0 \sim DP(\alpha G_0)$$ $$\beta | \beta_0, S_{\beta_0} \sim N(\beta_0,S_{\beta_0})$$ where, $G0 = N(theta| mu, sigma2)$. To complete the model specification, independent hyperpriors are assumed, $$\alpha | a_0, b_0 \sim Gamma(a_0,b_0)$$ $$\mu | \mu_b, S_b \sim N(\mu_b,S_b)$$ $$\sigma^{-2} | \tau_1, \tau_2 \sim Gamma(\tau_1/2,\tau_2/2)$$ The precision or total mass parameter, $alpha$, of the DP prior can be considered as random, having a gamma distribution, $Gamma(a0,b0)$, or fixed at some particular value. When $alpha$ is random the method described by Escobar and West (1995) is used. To let $alpha$ to be fixed at a particular value, set $a0$ to NULL in the prior specification.

Each of the parameters of the baseline distribution, $mu$ and $sigma2$ can be considered as random or fixed at some particular value. In the first case, a Mixture of Dirichlet Process is considered as a prior for the distribution of the random effects. To let $sigma2$ to be fixed at a particular value, set $tau1$ to NULL in the prior specification. To let $mu$ to be fixed at a particular value, set $mub$ to NULL in the prior specification.

The computational implementation of the model is based on the marginalization of the DP and on the use of MCMC methods for nonconjugate priors (see, MacEachern and Muller, 1998; Neal, 2000). Specifically, the algorithm 8 with m=1 of Neal (2000), is considered in the DPraschpoisson function. In this case, the full conditional distributions for the difficulty parameters and in the resampling step of random effects are generated through the Metropolis-Hastings algorithm with a IWLS proposal (see, West, 1985 and Gamerman, 1997).

The functionals parameters are sampled using the $epsilon$-DP approximation proposed by Muliere and Tardella (1998), with $epsilon$=0.01.

References

Escobar, M.D. and West, M. (1995) Bayesian Density Estimation and Inference Using Mixtures. Journal of the American Statistical Association, 90: 577-588.

Gamerman, D. (1997) Sampling from the posterior distribution in generalized linear mixed models. Statistics and Computing, 7: 57-68.

MacEachern, S. N. and Muller, P. (1998) Estimating mixture of Dirichlet Process Models. Journal of Computational and Graphical Statistics, 7 (2): 223-338.

Muliere, P. and Tardella, L. (1998) Approximating distributions of random functionals of Ferguson-Dirichlet priors. The Canadian Journal of Statistics, 26(2): 283-297.

Neal, R. M. (2000) Markov Chain sampling methods for Dirichlet process mixture models. Journal of Computational and Graphical Statistics, 9:249-265.

San Martin, E., Jara, A., Rolin, J.-M., and Mouchart, M. (2011) On the Bayesian nonparametric generalization of IRT-type models. Psychometrika (To appear)

West, M. (1985) Generalized linear models: outlier accomodation, scale parameter and prior distributions. In Bayesian Statistics 2 (eds Bernardo et al.), 531-558, Amsterdam: North Holland.

See Also

DPrandom, FPTrasch

Examples

## Not run: 
#     ####################################
#     # A simulated Data Set
#     ####################################
#       nsubject <- 200
#       nitem <- 40
#       
#       y <- matrix(0,nrow=nsubject,ncol=nitem)
#       dimnames(y) <- list(paste("id",seq(1:nsubject)), 
#                           paste("item",seq(1,nitem)))
# 
#       
#       ind <- rbinom(nsubject,1,0.5)
#       theta <- ind*rnorm(nsubject,1,0.25)+(1-ind)*rnorm(nsubject,3,0.25)
#       beta <- c(0,seq(-1,3,length=nitem-1))
#       true.cdf <- function(grid)
#       {
#          0.5*pnorm(grid,1,0.25)+0.5*pnorm(grid,3,0.25) 
#       }  
#       for(i in 1:nsubject)
#       {
#          for(j in 1:nitem)
#          {
#             eta<-theta[i]-beta[j]         
#             mean<-exp(eta)/(1+exp(eta))
#             y[i,j]<-rbinom(1,1,mean)
#          }
#       }
# 
#     # Prior information
# 
#       beta0 <- rep(0,nitem-1)
#       Sbeta0 <- diag(1000,nitem-1)
# 
#       prior <- list(alpha=1,
#                     tau1=6.02,
#                     tau2=2.02,
#                     mub=0,
#                     Sb=100,
#                     beta0=beta0,
#                     Sbeta0=Sbeta0)
# 
#     # Initial state
#       state <- NULL
# 
#     # MCMC parameters
# 
#       nburn <- 5000
#       nsave <- 5000
#       nskip <- 0
#       ndisplay<- 1000
#       mcmc <- list(nburn=nburn,
#                    nsave=nsave,
#                    nskip=nskip,
#                    ndisplay=ndisplay)
# 
#     # Fit the model
#       fit1 <- DPrasch(y=y,prior=prior,mcmc=mcmc, 
#                       state=state,status=TRUE,grid=seq(-1,5,0.01),
#                       compute.band=TRUE)
# 
#     # CDF estimate and truth
#       plot(fit1$grid,true.cdf(fit1$grid),type="l",lwd=2,col="red",
#            xlab=expression(theta),ylab="CDF")
#       lines(fit1$grid,fit1$cdf,lwd=2,col="blue")
#       lines(fit1$grid,fit1$cdf.l,lwd=2,col="blue",lty=2)
#       lines(fit1$grid,fit1$cdf.u,lwd=2,col="blue",lty=2)
# 
#     # Summary with HPD and Credibility intervals
#       summary(fit1)
#       summary(fit1,hpd=FALSE)
# 
#     # Plot model parameters 
#     # (to see the plots gradually set ask=TRUE)
#       plot(fit1,ask=FALSE)
#       plot(fit1,ask=FALSE,nfigr=2,nfigc=2)	
# 
#     # Extract random effects
#     
#       DPrandom(fit1)
#       plot(DPrandom(fit1))
#       DPcaterpillar(DPrandom(fit1))
# ## End(Not run)