For information and details about the predictor criticality measure, see:

• Azen, R., Budescu, D. V., and Reiser, B. (2001). Criticality of predictors in multiple regression. British Journal of Mathematical and Statistical Psychology, 54, pp 201-225.

This SAS program requires that you have an ASCII (plain text) data file with the variables arranged in the order Y, X₁, …, X_p.

The OUTPUT of this program includes the following information:

• Regression analysis using the input data file.
• For each fit criterion (AIC, R_adj², C_p):
  • The Best-Fitting-Models frequency distribution;
  • The criticality values and their 95% confidence intervals;
  • The weighted criticality values;
  • The dominance criticality values.

You will need to INPUT the following in the program:

• Name of the data file (an ASCII data file with variables arranged in the order Y, X₁, X₂, …, X_p).
• p: the number of predictors.
• B: the number of bootstrap runs.
• Type of predictors: either ‘f’ if the predictors are fixed, or ‘r’ if the predictors are random.
• A generating seed for the resamples: six digits (or 0 to use clock value).

THE PROGRAM:

  criticality.sas  (Click to download).