The HPCDM Procedure(Experimental)
ODS Graphics
Statistical procedures use ODS Graphics to create graphs as part of their output. ODS Graphics is described in detail in Chapter 21: Statistical Graphics Using ODS in SAS/STAT 14.1 User's Guide.
Before you create graphs, ODS Graphics must be enabled (for example, with the ODS GRAPHICS ON statement). For more information about enabling and disabling ODS Graphics, see the section "Enabling and Disabling ODS Graphics" in that chapter.
The overall appearance of graphs is controlled by ODS styles. Styles and other aspects of using ODS Graphics are discussed in the section "A Primer on ODS Statistical Graphics" in that chapter.
This section describes the use of ODS for creating graphics with the HPCDM procedure.
Note: If you request simulation of an aggregate loss sample of large size, either by specifying a large value for the NREPLICATES= option or by including a large number of replicates in the DATA= data set that you specify in conjunction with the EXTERNALCOUNTS statement, then it is recommended that you not request any plots, because creating plots that have large numbers of points can require a very large amount of hardware resources and can take a very long time. You can disable the generation of plots either by submitting the ODS GRAPHICS OFF statement before submitting the PROC HPCDM step or by specifying the PLOTS=NONE option in the PROC HPCDM statement. It is recommended that you request plots only when the sample size is less than 100,000.
ODS Graph Names
PROC HPCDM assigns a name to each graph that it creates by using ODS. You can use these names to selectively refer to the graphs. The names are listed in Table 4.4.
Table 4.4: ODS Graphics Produced by PROC HPCDM
|
ODS Graph Name |
Plot Description |
PLOTS= Option |
|---|---|---|
|
ConditionalDensityPlot |
Conditional density plot |
CONDITIONALDENSITY |
|
DensityPlot |
Probability density function plot |
DENSITY |
|
EDFPlot |
Empirical distribution function plot |
EDF |
Conditional Density Plot
The conditional density plot helps you visually analyze two or three regions of the compound distribution by displaying a density function estimate that is conditional on the values of the aggregate loss that fall in those regions. You can specify the region boundaries in terms of quantiles by using the LEFTQ= and RIGHTQ= suboptions of the PLOTS=CONDITIONALDENSITY option. This is especially useful if you want to see the distribution of aggregate loss values in the right- and left-tail regions.
If you specify the programming statements and the ADJUSTEDSEVERITY= symbol, then a separate set of conditional density plots are displayed for the aggregate adjusted loss.
Probability Density Function Plot
The probability density function (PDF) plot shows the nonparametric estimates of the PDF of the aggregate loss distribution. This plot includes histogram and kernel density estimates.
If you specify the programming statements and the ADJUSTEDSEVERITY= symbol, then a separate density plot is displayed for the aggregate adjusted loss.
Empirical Distribution Function Plot
The empirical density function (EDF) plot shows the nonparametric estimate of the cumulative distribution function of the aggregate loss distribution. You can specify the ALPHA= suboption of the PLOTS=EDF option to request that the upper and lower confidence limits be plotted for each EDF estimate. By default, the confidence interval is not plotted.
If you specify the programming statements and the ADJUSTEDSEVERITY= symbol, then a separate EDF plot is displayed for the aggregate adjusted loss.