topic Check categorical variable distribution in Statistical Procedures

Check categorical variable distribution

Ujjawal — Tue, 04 Aug 2015 21:04:11 GMT

I want to check categorical variable distribution over a period of time. For example, the distribution of variable district code has changed over a period of time ( Time 0 - Period of model Development, time1 - after one year).

I perform the following analysis for numeric variables.

Step 1 - Rank variable into 10 group (decile)

Step 2- (% of records based on variable in Scoring Sample (A) - % of records based on variable in Training Sample (B)) * In(A/ B)

Step 3 - Then sum up the scores in step2 on 10 groups

I'm little bit skeptical about the same analysis for categorical variables. I guess chi square analysis isn't correct technique to check it. Am i correct? What's the correct way?

Re: Check categorical variable distribution

Reeza — Tue, 04 Aug 2015 21:31:43 GMT

Take a look at the the options in proc freq, there's a test for trends over time.

Re: Check categorical variable distribution

Ujjawal — Wed, 05 Aug 2015 04:27:56 GMT

Which test?

Re: Check categorical variable distribution

Ksharp — Wed, 05 Aug 2015 13:02:43 GMT

This categorical variable is nominal or ordered ?

and how many levels does this variable have ?

If two level , you could try TRAND analysis.

tables var * time / trend cl ;

Otherwise , try MEASURE analysis.

tables var* Dose / measures cl;

Xia Keshan

Re: Check categorical variable distribution

Reeza — Wed, 05 Aug 2015 14:30:38 GMT

TREND test

Re: Check categorical variable distribution

Ksharp — Wed, 05 Aug 2015 14:46:10 GMT

Trend test is only for 2*N matrix , if I was right . while MEASURE(Association test) is for N*M matrix.

Re: Check categorical variable distribution

Reeza — Wed, 05 Aug 2015 14:50:16 GMT

OP mentioned 2*N - Time before and Time after measurements.