SAS Code for Performance of Model vs. Chance Model

Data defines the model by dint of genetic programming, producing the best decile table.

SAS Code for Performance of Model vs. Chance Model
Bruce Ratner, PhD

Live chat by Bold chat

data Model_CellCounts;
input ACTUAL PREDICTED Count @@;
datalines;
0 1 10 0 2 0
1 1 3 1 2 8
;
run;

proc freq data=Model_CellCounts;
table ACTUAL*PREDICTED /chisq sparse out=D;
weight Count;
run;

proc transpose data=D out=transp;
run;

data IMPROV;
retain MODEL_TCCR;
set transp;
drop _LABEL_;
if _NAME_=" ACTUAL" then delete;
if _NAME_=" PREDICTED " then delete;
if _NAME_="PERCENT" then CHANCE_TCCR=( ((col1+col2)**2)+((col3+col4)**2))/10000 ;
if _NAME_="COUNT" then MODEL_TCCR=((col1+col4)/sum(of col1-col4))/1;

IMPROV= ((MODEL_TCCR- CHANCE_TCCR)/CHANCE_TCCR);
if IMPROV=. then delete;
run;

proc print data=IMPROV;
var MODEL_TCCR CHANCE_TCCR IMPROV;
format CHANCE_TCCR MODEL_TCCR IMPROV percent8.2;
run;

For more information about this article, call Bruce Ratner at 516.791.3544 or 1 800 DM STAT-1; or e-mail at br@dmstat1.com.

Sign-up for a free GenIQ webcast: Click here.