topic Re: Fisher's - Taking long to run - 4x4 table with greater than 1000 sample size in New SAS User

Fisher's - Taking long to run - 4x4 table with greater than 1000 sample size

varatt90 — Tue, 09 May 2023 19:35:34 GMT

Hi,

I tried to run a fisher's test, as I have some cells with less than 5 counts, but my sample size is 6,000. As a result, the time to run the code is taking long.

Is there another test I could use or should I just wait it out?

Thank you

Re: Fisher's - Taking long to run - 4x4 table with greater than 1000 sample size

ballardw — Tue, 09 May 2023 19:51:53 GMT

How long is "long"?

Can you share the code that you are currently running? There are multiple procedures that use Fisher's test so we really need to know which one you are using.

Re: Fisher's - Taking long to run - 4x4 table with greater than 1000 sample size

varatt90 — Tue, 09 May 2023 19:55:15 GMT

Thank you for your response.

The code has been running for 45 min now.

My code is:

PROC FREQ DATA = Dataset;
TABLES Variable1*variable2/ FISHER;
RUN;

Re: Fisher's - Taking long to run - 4x4 table with greater than 1000 sample size

PaigeMiller — Tue, 09 May 2023 20:33:01 GMT

It's always a good idea, highly recommended, to read the documentation which says:

Note: PROC FREQ computes exact tests by using fast and efficient algorithms that are superior to direct enumeration. Exact tests are appropriate when a data set is small, sparse, skewed, or heavily tied. For some large problems, computation of exact tests might require a substantial amount of time and memory. Consider using asymptotic tests for such problems. Alternatively, when asymptotic methods might not be sufficient for such large problems, consider using Monte Carlo estimation of exact p-values. You can request Monte Carlo estimation by specifying the MC computation-option in the EXACT statement. See the section Computational Resources for more information.

Re: Fisher's - Taking long to run - 4x4 table with greater than 1000 sample size

varatt90 — Tue, 09 May 2023 20:40:14 GMT

Thank you, that worked! If anyone else is wondering, I was able to produce the F-test using the code:

PROC FREQ DATA = dataset;
TABLES variable1*variable2;
EXACT FISHER/ MC;
RUN;