The MCMC Procedure

 
ODS Table Names

PROC MCMC assigns a name to each table it creates. You can use these names to reference the table when using the Output Delivery System (ODS) to select tables and create output data sets. These names are listed in the following table. For more information about ODS, see Chapter 21, Statistical Graphics Using ODS.

Table 54.43 ODS Tables Produced in PROC MCMC

ODS Table Name

Description

Statement or Option

AutoCorr

Autocorrelation statistics for each parameter

Default

BurnInHistory

History of burn-in phase sampling

MCHISTORY=BRIEF | DETAILED

ConvergenceStatus

Optimization convergence status

PROPCOV=method(ITPRINT)

Corr

Correlation matrix of the posterior samples

STATS=CORR

Cov

Covariance matrix of the posterior samples

STATS=COV

DIC

Deviance information criterion

DIC

ESS

Effective sample size for each parameter

Default

MCSE

Monte Carlo standard error for each parameter

Default

Geweke

Geweke diagnostics for each parameter

Default

Heidelberger

Heidelberger-Welch diagnostics for each parameter

DIAGNOSTICS=HEIDEL

InputOptions

Optimization input table

PROPCOV=method(ITPRINT)

PostIntervals

Equal-tail and HPD intervals for each parameter

Default

IterHist

Optimization iteration history

PROPCOV=method(ITPRINT)

IterStop

Optimization results table

PROPCOV=method(ITPRINT)

NObs

Number of observations

Default

OptiEstimates

Parameter values after either optimization

PROPCOV=method(ITPRINT)

OptiCov

Covariance used in proposal distribution after optimization

PROPCOV=method(ITPRINT)

Parameters

Summary of the PARMS, BLOCKING, PRIOR, sampling method, and initial value specification

Default

ParametersInit

Parameter values after the tuning phase

INIT=PINIT

PosteriorSample

Posterior samples for each parameter

(for ODS output data set only)

PostSummaries

Basic posterior statistics for each parameter, including sample size, mean, standard deviation, and percentiles

Default

PredSummaries

Basic posterior statistics for each prediction

Default with any PREDDIST statement

PredIntervals

Equal-tail and HPD intervals for each prediction

Default with any PREDDIST statement

ProblemDescription

Optimization table

PROPCOV=method(ITPRINT)

REParameters

Random effect, subject variable, number of levels, and prior distribution of the random effect

Default with any RANDOM statement

Raftery

Raftery-Lewis diagnostics for each parameter

DIAGNOSTICS=RAFTERY

SamplingHistory

History of main phase sampling

MCHISTORY=BRIEF | DETAILED

TuneCov

Proposal covariance matrix (for continuous parameters) after the tuning phase

INIT=PINIT

TuneP

Proposal probability vector (for discrete parameters) after the tuning phase

INIT=PINIT and DISCRETE=GEO

TuningHistory

History of proposal distribution tuning

MCHISTORY=BRIEF | DETAILED