object.coda.bairt {bairt} | R Documentation |
The function object.coda create a mcmc.list object. With this is possible to study the chain using the coda packet.
## S3 method for class 'bairt' object.coda(mcmclist, parameter = "a", chain = 1, parts = NULL, ...)
mcmclist |
A mcmc.2pnob or mcmc.3pnob class object. |
parameter |
The parameter (a, b, c or theta) for graphing. |
chain |
The parameter's chain that will be graphed. |
parts |
Number of splits for MCMC chain. |
... |
Further arguments. |
The function object.coda create a mcmc.list object of the marginal chain selectionated. The marginal chain is splited in subchains determined by parts. The aim is represent parallel chains with different starting values (Beguin & Glas, 2001, p. 547).
A mcmc.list coda packet object.
Javier MartÃnez
A.A. Beguin, A, A. & Glas, C.A.W. (2001). MCMC Estimation and Some Model-Fit Analysis of Multidimensional IRT Models. Psychometrika, 66, 541-562.
as.mcmc.list
and as.mcmc
.
# data for model data("MathTest") # Only for the first 500 examinees of the data MathTest # Two-Parameter Normal Ogive Model model2 <- mcmc.2pnob(MathTest[1:500,], iter = 400, burning = 100) chain_a1 <- object.coda(model2, parameter = "a", chain = 1) coda::gelman.plot(chain_a1) coda::gelman.diag(chain_a1) plot(chain_a1) # For all examinees of the data MathTest # Three-Parameter Normal Ogive Model # selection of the prior for 5 response options cprior <- select.c.prior(5) model3 <- mcmc.3pnob(MathTest, iter = 3500, burning = 500, c.prior = cprior, parts = 3) chain_c1 <- object.coda(model3, parameter = "c", chain = 1) coda::gelman.plot(chain_c1) coda::gelman.diag(chain_c1) plot(chain_c1) ## End(Not run)