Read XLS with data starts on 3rd row, and column names on 2nd row

proc import file='C:\Users\procx\sample.xls'

        out=test dbms=xls replace;

        sheet=disposition;

        namerow=2;

        startrow=3;

        getnames=yes;

run;

Get around ‘WARNING: Fisher's exact test is not computed when the total sample size exceeds 32767.’

Fisher's Exact test can be memory and computation extensive when the sample size is large. For the large sample size problem, you can use Monte-Carlo estimates of the exact p-value to get around the issue. For example,

exact fisher /mc;

More details of the MC option can be found here -

http://support.sas.com/documentation/cdl/en/statug/67523/HTML/default/viewer.htm#statug_freq_syntax03.htm#statug.freq.freqexfisher