The QUANTLIFE Procedure

PROC QUANTLIFE Statement

  • PROC QUANTLIFE <options> ;

The PROC QUANTLIFE statement invokes the QUANTLIFE procedure. Table 94.1 summarizes the options available in this statement.

Table 94.1: Options Available in the PROC QUANTLIFE Statement

Option

Description

Data Set Options

DATA=

Specifies the input SAS data set

OUTBOOTEST=

Creates an output SAS data set for parameter estimates from resampled data sets

Basic Options

ALPHA=

Specifies the confidence level

CI=

Specifies a resampling method for computing confidence interval and test statistics

LOG

Requests log transformation of the response

METHOD=

Specifies a method to fit quantile regression

NAMELEN=

Specifies the length of effect names

NREP=

Specifies the number of replications

SEED=

Specifies the seed for the random number generator

PLOTS=

Specifies the plots to be produced by using ODS Graphics

Computational Options

GRIDSIZE=

Specifies a step size for the grid for computing regression quantiles

INITTAU=

Specifies the first quantile level for computing regression quantiles

KAPPA=

Specifies the step-length parameter for the interior point algorithm

MAXIT=

Specifies the maximum number of iterations for the interior point algorithm

NTHREADS=

Specifies the number of threads for the computation

TOLERANCE=

Specifies the convergence criterion of the interior point algorithm


You can specify the following options in the PROC QUANTLIFE statement.

ALPHA=value

specifies the confidence level for the regression parameters. The value must be between 0 and 1. By default, ALPHA=0.05, which corresponds to a 95% confidence interval.

CI=EW | PW | NONE

requests the method for computing confidence intervals for regression parameters. In addition to confidence intervals, the QUANTLIFE procedure also computes standard errors, t values, and p-values for regression parameters. You can suppress these computations by specifying CI=NONE. The QUANTLIFE procedure provides two resampling methods for computing confidence intervals, the exponentially weighted (EW) method and the pairwise (PW) resampling method. For more information, see the section Confidence Interval. By default, CI=EW, which requests the exponentially weighted method.

DATA=SAS-data-set

specifies the SAS-data-set that the QUANTLIFE procedure uses. By default, the most recently created SAS-data-set is used.

GRIDSIZE=value

specifies the step size for computing regression quantiles. The value must be between 0 and 1. For more information, see the section Details: QUANTLIFE Procedure.

INITTAU=value

specifies the first quantile level for computing regression quantiles. The value must be between 0 and 1. For more information, see the section Details: QUANTLIFE Procedure.

KAPPA=value

specifies the step-length parameter for the interior point algorithm. The value must be between 0 and 1. The interior point method that the QUANTLIFE procedure uses is identical to the interior point method that the QUANTREG procedure uses. For more information, see the section Interior Point Algorithm in Chapter 95: The QUANTREG Procedure. By default, KAPPA=0.99995.

LOG

requests that a log transformation of the response variable be performed before the model is fitted.

MAXIT=n

specifies the maximum number of iterations for the interior point algorithm. By default, MAXIT= 1000.

METHOD=KM | NA 

requests the method to use to estimate the regression parameters. METHOD=KM specifies the Kaplan-Meier-type method (see the section Kaplan-Meier-Type Estimator for Censored Quantile Regression) and METHOD=NA specifies the Nelson-Aalen-type method (see the section Nelson-Aalen-Type Estimator for Censored Quantile Regression). By default, METHOD=KM.

NAMELEN=n

specifies the length of effect names in tables and output data sets to be n characters, where n is a value between 20 and 200. By default, NAMELEN=20.

NREP=n

specifies the number of replications to draw in the resampling method. By default, NREP=200.

NTHREADS=n

specifies the number of threads for analytic computations and overrides the SAS system option THREADS | NOTHREADS. If you do not specify the NTHREADS= option or if you specify NTHREADS=0, the number of threads is determined based on the data size and the number of CPUs on the host where the analytic computations execute.

OUTBOOTEST=SAS-data-set

creates a data set to contain the parameter estimates from the resampled data sets. For a detailed description of the contents of the OUTBOOTEST= data set, see the section OUTBOOTEST= Output Data Set.

PLOTS =(plot-request < …plot-request >)

requests various plots.

When you specify one plot-request, you can omit the parentheses around it.

ODS Graphics must be enabled before plots can be requested. For example:

ods graphics on;

proc quantlife plots=survival;
   model y=x1;
run;

ods graphics off;

For more information about enabling and disabling ODS Graphics, see the section Enabling and Disabling ODS Graphics in Chapter 21: Statistical Graphics Using ODS.

You can specify one or more of the following plot-requests:

ALL

creates all appropriate plots.

NONE

suppresses all the plots in the procedure. Specifying this option is equivalent to disabling ODS Graphics for the entire procedure.

QUANTILE

plots the estimated quantile function for each combination of covariate values in the COVARIATES= data set that is specified in the BASELINE statement. If the COVARIATES= data set is not specified, then the estimated quantile function is plotted for the reference set of covariate values, which consists of reference levels for the CLASS variables and average values for the continuous variables. When the estimated quantile function is not monotonic, the quantile function (Chernozhukov, Fernandez-Val, and Galichon 2009) is rearranged to make it monotonic and then plotted.

QUANTPLOT </ UNPACK >

plots the regression quantile process. The estimated coefficient of each specified covariate effect is plotted as a function of the quantile level. You can use the UNPACK option to create individual process plots.

SURVIVAL

plots the estimated survival function for each combination of covariate values in the COVARIATES= data set that is specified in the BASELINE statement. If the COVARIATES= data set is not specified, then the estimated survival function is plotted for the reference set of covariate values, which consists of reference levels for the CLASS variables and average values for the continuous variables.

SEED=number

specifies a positive integer to start the pseudorandom number generator. The default is a value that is generated from reading the time of day from the computer’s clock. However, to duplicate the results under identical situations, you must specify the same seed in subsequent runs of the QUANTLIFE procedure. The seed information is displayed in the "Model Information" table.

TOLERANCE=value

specifies the tolerance for the convergence criterion of the interior point algorithm. Both the QUANTLIFE procedure and the QUANTREG procedure use the duality gap as the convergence criterion. For more information, see Interior Point Algorithm: Interior Point Algorithm. By default, TOLERANCE=1E–8.