The SURVEYPHREG Procedure

Displayed Output

If you use the NOPRINT option in the PROC SURVEYPHREG statement, the procedure does not display any output. Otherwise, PROC SURVEYPHREG displays results of the analysis in a collection of tables.

Model Information

The "Model Information" table displays the two-level name of the input data set, the name and label of the failure time variable, the name and label of the censoring variable and the values that indicate censored times, the model, the name and label of the FREQ variable, the name and label of the WEIGHT variable, the name and label of the STRATA variables, the name and label of the CLUSTER variables, and the method of handling ties in the failure time for the Cox model. The ODS name of the "Model Information" table is ModelInfo.

Number of Observations

The "Number of Observations" table displays the number of observations that are read and used, the sum of frequencies read and used, the sum of weights read and used, and the weighted sum of frequencies that are read and used in the analysis. The ODS name of the "Number of Observations" table is NObs.

Summary of the Number of Event and Censored Values

The "Summary of the Number of Event and Censored Values" table displays the number of events and censored values. The ODS name of the "Summary of the Number of Event and Censored Values" table is CensoredSummary.

Summary of the Weighted Number of Event and Censored Values

The "Summary of the Weighted Number of Event and Censored Values" table displays the weighted number of events and censored values. The ODS name of the "Summary of the Weighted Number of Event and Censored Values" table is WeightedCensoredSummary.

Class Level Information

The "Class Level Information" table is displayed when there are CLASS variables in the model. The table lists the categories of every CLASS variable that is used in the model and the corresponding design variable values. The ODS name of the "Class Level Information" table is ClassLevelInfo.

Design Summary Table

The "Design Summary" table provides information about the sample design. The table displays the total number of strata that are read and used, and the total number of clusters read and used. The table is displayed only if you specify a STRATA or CLUSTER statement. The ODS name of the "Design Summary" table is DesignSummary.

Stratum Information Table

If you specify the LIST option in the STRATA statement, PROC SURVEYPHREG displays a "Stratum Information" table. The ODS name of the "Stratum Information Table" is StrataInfo. This table provides the following information for each stratum:

  • Stratum Index, which is a sequential stratum identification number

  • STRATA variables, which list the levels of STRATA variables for the stratum

  • Number of Observations, which is the number of observations used in the stratum

  • Population Total for the stratum, if you specify the TOTAL= option

  • Sampling Rate for the stratum, if you specify the TOTAL= or RATE= option. If you specify the TOTAL= option, the sampling rate is based on the number of valid observations in the stratum.

  • Number of Clusters, which is the number of clusters in the stratum, if you specify a CLUSTER statement

Convergence Status

The "Convergence Status" table displays the convergence status of the optimization routine. The procedure displays this table only when you specify the NLOPTIONS statement. The ODS name of the "Convergence Status" table is ConvergenceStatus.

Model Fit Statistics

The "Model Fit Statistics" table displays the values of –2 log likelihood and the AIC for the null model and the fitted model. The ODS name of the "Model Fit Statistics" table is FitStatistics.

Testing Global Null Hypothesis: BETA=0

The "Testing Global Null Hypothesis: BETA=0" table displays results of the likelihood ratio test and the Wald test for testing the hypothesis that all parameters are zero. The ODS name of the "Testing Global Null Hypothesis: BETA=0" table is GlobalTests.

Analysis of Maximum Likelihood Estimates

The "Analysis of Maximum Likelihood Estimates" table displays the denominator degrees of freedom, which is computed as described in the section Degrees of Freedom; the maximum likelihood estimate of the parameter; the estimated standard error, computed as the square root of the corresponding diagonal element of the estimated covariance matrix; the t statistic, computed as the parameter estimate divided by the standard error; the p-value of the t statistic with respect to a t distribution with denominator degrees of freedom; and the hazard ratio estimate. The t confidence limits for the parameter estimates and estimated hazard ratios are displayed if you specify the CLPARM or RISKLIMITS option in the MODEL statement. You can specify the DF=NONE option in the MODEL statement to request p-values and confidence intervals from a standard normal distribution. If you specify the VARRATIO=ALL | MODEL | IND option in the MODEL statement, then the variance ratios for model or independence (or both) are displayed. If you specify the SERATIO=ALL | MODEL | IND option in the MODEL statement, then the standard error ratios for model or independence (or both) are displayed.

The ODS name of the "Analysis of Maximum Likelihood Estimates" table is ParameterEstimates.

Covariance Matrix

The "Covariance Matrix" table is displayed if you specify the COVB option in the MODEL statement. The table contains the estimated covariance matrix for the parameter estimates. The ODS name of the "Covariance Matrix" table is CovB.

Hessian Matrix

The "Hessian Matrix" table is displayed if you specify the HESS option in the MODEL statement. The table contains the Hessian matrix that is evaluated at the estimated regression parameters. The ODS name of the "Hessian Matrix" table is Hessian.

Inverse Hessian Matrix

The "Inverse Hessian Matrix" table is displayed if you specify the INVHESS option in the MODEL statement. The table contains the inverse of the Hessian matrix evaluated at the estimated regression parameters. The ODS name of the "Inverse Hessian Matrix" table is InvHessian.

Variance Estimation Table

The "Variance Estimation" table provides the following information:

  • Method, which is the variance estimation method—Taylor Series, Balanced Repeated Replication, or Jackknife

  • Replicate Weights input data set name, if you provide replicate weights with a REPWEIGHTS statement

  • Number of Replicates, for VARMETHOD=BRR or VARMETHOD=JACKKNIFE

  • Hadamard Data Set name, if you specify the HADAMARD= method-option for VARMETHOD=BRR

  • Fay Coefficient, if you specify the FAY method-option for VARMETHOD=BRR

  • Missing Values Included (MISSING), if you specify the MISSING option

  • Missing Values Included (NOMCAR), if you specify the NOMCAR option

  • Missing Values Excluded, if you have missing values and you do not specify the NOMCAR option

The ODS name of the "Variance Estimation" table is VarianceEstimation.

Hadamard Matrix

If you specify the PRINTH method-option for VARMETHOD=BRR , PROC SURVEYPHREG displays the Hadamard matrix that is used to construct replicates for BRR variance estimation. If you provide a Hadamard matrix with the HADAMARD= method-option for VARMETHOD=BRR but the procedure does not use the entire matrix, the procedure displays only the rows and columns that are actually used to construct replicates. The ODS name of the "Hadamard Matrix" table is HadamardMatrix.

Maximum Likelihood Estimates for Replicate Samples

If you specify the DETAILS method-option for VARMETHOD=BRR or the DETAILS method-option for VARMETHOD=JACKKNIFE , PROC SURVEYPHREG displays the "Maximum Likelihood Estimates for Replicate Samples" table. The Replicate Number column displays the replication number, the Replicate Weight column displays the name of the replicate weight variable, and the Status column displays the convergence status. The replicate number for the full sample is set to 0. If you do not specify replicate weights, then PROC SURVEYPHRG uses default names to identify the replicate weights. For more information, see Replicate Weights Output Data Set. The convergence status is 1 if the maximum likelihood estimates are available for a replicate sample and 0 otherwise. If the maximum likelihood estimates are not available for a replicate sample, then the parameter estimates are set to missing for that replicate. The ODS name of the "Maximum Likelihood Estimates for Replicate Samples" table is RepEstimates.