CRITERIA Subcommand (BAYES RELATED command)
CRITERIA
controls the statistical criteria used in building the Bayesian Related
Sample Inference models. The way in which these criteria are used depends on the method specified on
the METHOD
subcommand. The default criteria are noted in the description of each
keyword below.
CILEVEL = value. Specify the significance level for computing credible intervals. Must
be a single double value ∈(0,100)
. The default level is 95%.
METHOD. Specify the numerical method to estimate the integral. Valid values include AGL and MCA. AGL calls the adaptive Gauss-Lobatto quadrature approach (the default setting). MCA calls the Monte Carlo approximation approach.
SEED = integer. Specify a random seed set for the MCA method, with a random seed assigned
by default. Only in effect when METHOD = MCA. You can set a seed by
specifying a positive integer ∈[1,2147483647]
.
TOL = value. Specify the tolerance value for the numerical methods.
0.000001
is the default setting. Only in effect when METHOD = AGL
.
Must be one value ϵ(0, 10-3]
.
MAXITER = integer. Specify the maximum number of iterations the AGL method can go through
in its computations. The integer 2000
is the default setting. You must specify
a positive integer. Valid when METHOD = AGL
.
MCSAMPLES = integer. Specify the number of points that are sampled for the Monte Carlo
approximation. Only in effect when METHOD = MCA. 10,000 is the default
setting. The value must be an integer ∈[103,106]
.