PLOT Subcommand (QUANTILE REGRESSION command)
The PLOT
subcommand controls the plots to be output.
- MAX_CATEGORIES = {10**} {integer}
- The optional keyword controls the number of categories or combinations of
categories in a mixed effect to be plotted within the interaction of one covariate and one or two
factors. The value must be a single positive integer with
10
as the default setting. When the specified integer is larger than the number of categories or combinations, the prediction plots are created for all categories and combinations. - PREDICTED_BY_OBSERVED = {FALSE**} {TRUE}
- The optional keyword controls whether to create the predicted versus observed values plot. FALSE is the default setting. When TRUE is specified, a single plot is created that contains the points with different colored dots that represent different quantiles.
- PARAM_EST = {50**} {integer} {ALL}
- The optional keyword controls the plots for the parameter estimates with
their confidence intervals controlled by /CRITERIA CILEVEL. When a numeric value
is specified, it must be a single non-negative integer with
50
as the default setting. With the default setting, the first 50 non-redundant parameters are plotted. All parameters are plotted when the specified integer is larger than the number of non-redundant parameters contained in the model. When the string “ALL
”is specified, all parameters in effect are plotted. The keyword takes effect only when multiple values are specified for /CRITERIA QUANTILE. No plots are created whenPARAM_EST = 0
or a single quantile is specified.