prioriPlot {eiwild} | R Documentation |
prioriPlot simulates the beta_i^rc values at first level given specific parameters at hyperpriori level
prioriPlot(pars, which, cols, alphaSample = 10000, betaSample = 300, plot = TRUE, ...)
pars |
list of parameters for hyperpriori. if which="gamma" then parameter has to be a list with shape and rate as parameters if which="expo" then parameter has to be a list with only lam |
which |
specified priori. "gamma" or "expo" |
cols |
integer specifying how many columns the RxC-Table should have |
alphaSample |
integer specifying the number of times new alpha-values are drawn |
betaSample |
integer specifying the number of times betas will be drawn for each alpha-value |
plot |
logical TRUE/FALSE if histogram should be plotted |
... |
additional arguments for "hist" function |
Calculation is made via the marginal beta distribution
function structure:
"gamma"
choose
one parameter for every alpha_rc-parameter or a two
matrices of parameters specifying lambda's for every
alpha_rc-parameter
"expo"
choose one parameter
for every alpha_rc-parameter or a one matrix of parameters
specifying lambda's for every alpha_rc-parameter
nested list
with each element containing another
list
. First level are rows and second level are
columns per row.
## Not run: test1 <- prioriPlot(list(shape=4,rate=2), "gamma",cols=4) str(test1) pars <- list(shape=matrix(1:9,3,3),rate=matrix(9:1,3,3)) test2 <- prioriPlot(pars, "gamma",breaks=100) test3 <- prioriPlot(list(shape=8,rate=2),"gamma",breaks=100,cols=3) pars4 <- list(shape=matrix(c(6,6,6),1,3), rate=matrix(c(4,4,4),1,3)) test4 <- prioriPlot(pars4, "gamma",breaks=100) pars5 <- list(lam=2) test5 <- prioriPlot(pars5, "expo",cols=4, breaks=100) pars6 <- list(lam=matrix(1:9,3,3)/100) test6 <- prioriPlot(pars6, "expo", breaks=25, col=grey(0.8)) # example for 3x4-table set.seed(568) pars7 <- list(shape=matrix(sample(1:20,12), 3,4), rate=matrix(sample(1:20,12),3,4)) test7 <- prioriPlot(pars7, "gamma",breaks=50) ## End(Not run)