print.sienaBayesFit {RSienaTest} | R Documentation |
print
and summary
methods for sienaBayesFit
objects,
and further functions for interpretation of results.
## S3 method for class 'sienaBayesFit' print(x, nfirst=NULL, ...) ## S3 method for class 'sienaBayesFit' summary(object, nfirst=NULL, ...) ## S3 method for class 'summary.sienaBayesFit' print(x, nfirst=NULL, ...) shortBayesResults(x, nfirst=NULL) plotPostMeansMDS(x, pmonly=0, nfirst=NULL, ...)
object |
An object of class |
x |
An object of class |
nfirst |
The first element of the MCMC chain used for calculating
properties of the chain; i.e., the first element for which
it is assumed that convergence has occurred. If |
pmonly |
if |
... |
For extra arguments. At present used only for
further specification of the plot made by |
The function print.sienaBayesFit
prints results (estimated posterior means,
posterior standard deviations, and percentiles) of a Bayesian analysis, as resulting
from function sienaBayes
.
The function summary.sienaBayesFit
prints more extensive results,
including groupwise results.
The columns "cred. from" and "cred. to" give the left and right end points
of the estimated 95 percent credibility interval for the parameters, and the
column "p" gives the estimated posterior probability that the parameter is
greater than 0.
The function shortBayesResults
produces a data frame containing
posterior means and standard deviations, and 95 percent percentiles.
This is meant to be a component for use in other functions.
Of this data frame, variables "name" to "effectName" are as in a
sienaEffects
object;
variables "postMeanGlobal", "postSdGlobal", "cFrom", "cTo" are the posterior
mean, posterior standard deviation, and lower and upper boundaries of the
95 percent posterior interval for the global parameters (mu
and
eta
);
variables "postSdBetween", "cSdFrom", "cSdTo" are the posterior mean,
and similar percentiles, for the global between-group standard deviations
(sigma
) of the parameters that are randomly varying between groups.
The function plotPostMeansMDS
makes an MDS plot of the posterior means.
This can be useful for detection of outliers.
Its return value is the matrix of two-dimensional MDS coordinates.
It is advisable to employ the first
parameter; looking at trace plots
will be helpful to estimate the number of runs from which convergence
has occurred; this may well be different from the default value x$nwarm+1
.
Function siena.table
can also be used to make html or LaTeX
tables of posterior means and standard deviations of a
sienaBayesFit
object.
Tom Snijders
See also http://www.stats.ox.ac.uk/~snijders/siena/
For the MDS method used in plotPostMeansMDS
: see p. 308 of
W.N. Venables and B.D. Ripley (2002),
Modern Applied Statistics with S. Fourth edition. Springer.
Group1 <- sienaDependent(array(c(N3401, HN3401), dim=c(45, 45, 2))) Group3 <- sienaDependent(array(c(N3403, HN3403), dim=c(37, 37, 2))) Group4 <- sienaDependent(array(c(N3404, HN3404), dim=c(33, 33, 2))) Group6 <- sienaDependent(array(c(N3406, HN3406), dim=c(36, 36, 2))) dataset.1 <- sienaDataCreate(Friends = Group1) dataset.3 <- sienaDataCreate(Friends = Group3) dataset.4 <- sienaDataCreate(Friends = Group4) dataset.6 <- sienaDataCreate(Friends = Group6) FourGroups <- sienaGroupCreate( list(dataset.1, dataset.3, dataset.4, dataset.6)) FourEffects <- getEffects(FourGroups) FourEffects <- includeEffects(FourEffects, transTrip) FourEffects <- setEffect(FourEffects, density, random=TRUE) FourEffects <- setEffect(FourEffects, recip, random=TRUE) print(FourEffects, includeRandoms=TRUE) # Note this also shows the "randomEffects" column # and the dimensions of the objects for specifying the priors. print(FourEffects, includeRandoms=TRUE, dropRates=TRUE) # Note this does not show the rate effects. FourAlgo <- sienaAlgorithmCreate(projname = "FourGroups", maxlike=TRUE, seed=123) ## Not run: bayes.model <- sienaBayes(FourAlgo, data = FourGroups, effects = FourEffects, nprewarm=10, nwarm=10, nmain=25, nrunMHBatches=10) bayes.model # If one would assume that converge was reached by iteration 20: print(bayes.model, nfirst=20) summary(bayes.model) shortBayesResults(bayes.model) # In the following, "main" is used as ... : plotPostMeansMDS(bayes.model, main="MDS of posterior means") siena.table(bayes.model, type="html", d=3) ## End(Not run)