topic Re: Gini Calculation in SAS Data Science

Gini Calculation

chemicalab — Wed, 18 Sep 2013 10:39:48 GMT

Hi all,

Any ideas on how the gini measure of diversity is calculated when creating bins through the interactive binning node? cause i see it alters when i either merge or spliit bins

Thank you in advance

Re: Gini Calculation

jwexler — Thu, 19 Sep 2013 15:11:14 GMT

Hi as you may already know, the Gini statistic tests for differences in bin values with respect to the target (events vs non-events). As Gini changes in the Interactive Binning node, you get an idea of the validity of your bin cutoffs.

You can find this equation in the EM reference guide in the Interactive Binning node section.

Thanks,

Jonathan

Re: Gini Calculation

chemicalab — Thu, 19 Sep 2013 16:31:09 GMT

Ok i think i am almost there, but how would the formula look like in the following example?

Value	Group	Cutoff	Event count	Non event count	Total	Event rate
MISSING	1		0	0	0	0
AGE <25	2	25.5	517	233	750	68.9%
25.5<AGE<31	3	31	365	281	646	56.5%
31.5<AGE<40	4	31.5	349	451	800	43.6%
40<AGE	5		269	535	804	33.5%

Thank you in advance

Re: Gini Calculation

jwexler — Tue, 24 Sep 2013 21:06:41 GMT

Hi, thanks to one of my R&D colleagues, he wrote up a quick solution using IML, which should match your results in EM.

proc iml;

event={517,365,349,269};

nonevent={233,281,451,535};

sum1=0;

do i=2 to 4;

sum2=0;

im1=i-1;

do j=1 to im1;

sum2=sum2+nonevent;

end;

sum1=sum1 + event*sum2;

end;

sum3=0;

do i=1 to 4;

sum3 = sum3 + event*nonevent;

end;

gini = 100*(1 - (2*sum1 + sum3)/ (1500*1500)) ;

print sum1 sum2 sum3 gini;

quit;

524016

965

524340

30.1168

Re: Gini Calculation

chemicalab — Wed, 25 Sep 2013 07:16:25 GMT

Thank you very much for your help