The RELIABILITY Procedure
- Overview
-
Getting Started
Analysis of Right-Censored Data from a Single PopulationWeibull Analysis Comparing Groups of DataAnalysis of Accelerated Life Test DataWeibull Analysis of Interval Data with Common Inspection ScheduleLognormal Analysis with Arbitrary CensoringRegression ModelingRegression Model with Nonconstant ScaleRegression Model with Two Independent VariablesWeibull Probability Plot for Two Combined Failure ModesAnalysis of Recurrence Data on RepairsComparison of Two Samples of Repair DataAnalysis of Interval Age Recurrence DataAnalysis of Binomial DataThree-Parameter WeibullParametric Model for Recurrent Events DataParametric Model for Interval Recurrent Events Data -
SyntaxPrimary StatementsSecondary StatementsGraphical Enhancement StatementsPROC RELIABILITY StatementANALYZE StatementBY StatementCLASS StatementDISTRIBUTION StatementEFFECTPLOT StatementESTIMATE StatementFMODE StatementFREQ StatementINSET StatementLOGSCALE StatementLSMEANS StatementLSMESTIMATE StatementMAKE StatementMCFPLOT StatementMODEL StatementNENTER StatementNLOPTIONS StatementPROBPLOT StatementRELATIONPLOT StatementSLICE StatementSTORE StatementTEST StatementUNITID Statement
-
DetailsAbbreviations and NotationTypes of Lifetime DataProbability DistributionsProbability PlottingNonparametric Confidence Intervals for Cumulative Failure ProbabilitiesParameter Estimation and Confidence IntervalsRegression Model Statistics Computed for Each Observation for Lifetime DataRegression Model Statistics Computed for Each Observation for Recurrent Events DataRecurrence Data from Repairable SystemsODS Table NamesODS Graphics
- References
Secondary Statements
You can specify the following statements in conjunction with the primary statements listed previously. These statements are used to modify the behavior of the primary statements or to specify additional variables.
BY
variables ;
CLASS
variables ;
DISTRIBUTION
distribution-name ;
EFFECTPLOT
<plot-type <(plot-definition-options)>> </ options> ;
ESTIMATE
<'label'> estimate-specification </ options> ;
FMODE
keyword = variable(’value1’ …’valuen’) ;
FREQ
variable ;
INSET
keyword-list </ options> ;
LSMEANS
<model-effects> </ options> ;
LSMESTIMATE
model-effect <'label'> values <divisor=n> <, …<'label'> values <divisor=n>> </ options> ;
MAKE
’table’ OUT=SAS-data-set < options> ;
NENTER
variable ;
NLOPTIONS
<options> ;
SLICE
model-effect </ options> ;
STORE
<OUT=>item-store-name </ LABEL='label'> ;
TEST
<model-effects> </ options> ;
UNITID
variable ;
The EFFECTPLOT, ESTIMATE, LSMEANS, LSMESTIMATE, SLICE, STORE, and TEST statements are used to provide further analysis of regression models that are fit by using a MODEL statement and are common to many SAS/STAT procedures. Summary descriptions of functionality and syntax for these statements appear after the PROC RELIABILITY statement in alphabetical order, and full documentation about them is available in Chapter 19: Shared Concepts and Topics in SAS/STAT 12.1 User's Guide.
You can use the STORE statement to store the results of fitting a regression model with a MODEL statement for later analysis with the SAS/STAT procedure PROC PLM, if you have SAS/STAT software installed at your site.
The BY statement specifies variables in the input data set that are used for BY processing. A separate analysis is performed for each group of observations defined by the levels of the BY variables. The input data set must be sorted in order of the BY variables.
The CLASS statement specifies variables in the input data set that serve as indicator, dummy, or classification variables in the MODEL statement.
The DISTRIBUTION statement specifies a probability distribution name for those statements that require a probability distribution for proper operation (the ANALYZE, PROBPLOT, MODEL, and RELATIONPLOT statements). If you do not specify a distribution with the DISTRIBUTION statement, the normal distribution is used.
The FMODE statement specifies what failure-mode data to include in the analysis of data. Use this statement in conjunction with the ANALYZE, MODEL, PROBPLOT, or RELATIONPLOT statement.
The FREQ statement specifies a variable that provides frequency counts for each observation in the input data set.
The INSET statement specifies what information is printed in the inset box created by the PROBPLOT or MCFPLOT statement. The INSET statement also controls the appearance of the inset box.
The MAKE statement creates a SAS data set from any of the tables produced by the procedure. You specify a table and a SAS data set name for the data set you want to create. There is a unique table name that identifies each table printed; see the tables in the section MAKE Statement.
The NENTER statement specifies interval-censored data having a special structure; these data are called readout data. Use the NENTER statement in conjunction with the FREQ statement.
The NLOPTIONS statement enables you to control aspects of the nonlinear optimizations used for maximum likelihood estimation of the parameters of the three-parameter Weibull distribution in the ANALYZE and PROBPLOT statements, and of parametric models for recurrent events data in the MODEL statement.
The UNITID statement specifies a variable in the input data set that is used to identify each individual unit in an MCFPLOT statement.