The HPSEVERITY Procedure

Displayed Output

The HPSEVERITY procedure optionally produces displayed output by using the Output Delivery System (ODS). All output is controlled by the PRINT= option in the PROC HPSEVERITY statement. Table 23.17 relates the ODS tables to PRINT= options.

Table 23.17: ODS Tables Produced in PROC HPSEVERITY

ODS Table Name

Description

Option

AllFitStatistics

Statistics of fit for all the distribution models

PRINT=ALLFITSTATS

ConvergenceStatus

Convergence status of parameter estimation process

PRINT=CONVSTATUS

DescStats

Descriptive statistics for the response variable

PRINT=DESCSTATS

DistributionInfo

Distribution information

PRINT=DISTINFO

EstimationDetails

Details of the estimation process for all the distribution models

PRINT=ESTIMATIONDETAILS

InitialValues

Initial parameter values and bounds

PRINT=INITIALVALUES

IterationHistory

Optimization iteration history

PRINT=NLOHISTORY

ModelSelection

Model selection summary

PRINT=SELECTION

OptimizationSummary

Optimization summary

PRINT=NLOSUMMARY

ParameterEstimates

Final parameter estimates

PRINT=ESTIMATES

PerformanceInfo

Execution environment information that pertains to the computational performance

Default

RegDescStats

Descriptive statistics for the regression effects that do not contain a CLASS variable

PRINT=DESCSTATS

StatisticsOfFit

Statistics of fit

PRINT=STATISTICS

Timing

Timing information for various computational stages of the procedure

DETAILS (PERFORMANCE statement)

TurnbullSummary

Turnbull EDF estimation summary

PRINT=ALL


If you do not specify the PRINT= option, then by default PROC HPSEVERITY produces ModelSelection, PerformanceInfo, ConvergenceStatus, OptimizationSummary, StatisticsOfFit, and ParameterEstimates ODS tables.

The following describes the content that each table displays:

AllFitStatistics (PRINT=ALLFITSTATS)

displays the comparison of all the statistics of fit for all the models in one table. The table does not include the models whose parameter estimation process does not converge. If all the models fail to converge, then this table is not produced. If the table contains more than one model, then the best model according to each statistic is indicated with an asterisk (*) in that statistic’s column.

ConvergenceStatus (PRINT=CONVSTATUS)

displays the convergence status of the parameter estimation process.

DescStats (PRINT=DESCSTATS)

displays the descriptive statistics for the response variable.

DistributionInfo (PRINT=DISTINFO)

displays the information about all the candidate distribution. It includes the name, the description, the number of distribution parameters, and whether the distribution is valid for the specified modeling task.

EstimationDetails (PRINT=ESTIMATIONDETAILS)

displays the comparative details of the estimation process that is used to fit each candidate distribution. If you specify the DETAILS option in the PERFORMANCE statement, then this table contains a column that indicates the time taken to estimate each candidate model.

InitialValues (PRINT=INITIALVALUES)

displays the initial values and bounds used for estimating each model.

IterationHistory (PRINT=NLOHISTORY)

displays the iteration history of the nonlinear optimization process used for estimating the parameters.

ModelSelection (PRINT=SELECTION)

displays the model selection table. The table shows the convergence status of each candidate model, and the value of the selection criterion along with an indication of the selected model.

OptimizationSummary (PRINT=NLOSUMMARY)

displays the summary of the nonlinear optimization process used for estimating the parameters.

ParameterEstimates (PRINT=ESTIMATES)

displays the final estimates of parameters. The estimates are not displayed for models whose parameter estimation process does not converge.

PerformanceInfo

displays information about the execution mode. For single-machine mode, the table displays the number of threads that are used. For distributed mode, the table displays the grid mode (symmetric or asymmetric), the number of compute nodes, and the number of threads per node. PROC HPSEVERITY produces this table by default.

RegDescStats (PRINT=DESCSTATS)

displays the descriptive statistics for the regression effects in the SCALEMODEL statement that do not contain a CLASS variable.

StatisticsOfFit (PRINT=STATISTICS)

displays the statistics of fit for each model. The statistics of fit are not displayed for models whose parameter estimation process does not converge.

Timing (DETAILS option in the PERFORMANCE statement

displays elapsed times (absolute and relative) for the main tasks of the procedure. PROC HPSEVERITY produces this table when you specify the DETAILS option in the PERFORMANCE statement,

TurnbullSummary (PRINT=ALL)

displays the summary of Turnbull’s estimation process if Turnbull’s method is used for computing EDF estimates. The summary includes whether the nonlinear optimization converged, the number of iterations, the maximum absolute relative error, the maximum absolute reduced gradient, and whether the final estimates are maximum likelihood estimates. This table is produced only if you specify PRINT=ALL and Turnbull’s method is used for computing EDF estimates.