topic Re: How to do multi-imputation based on Bernoulli method in Statistical Procedures

How to do multi-imputation based on Bernoulli method

sunny_0514 — Sat, 11 Nov 2023 09:14:26 GMT

I have a set of data that records the results of the examination of patients in the control group and the experimental group, divided into healed and non-union, but there are some missing values, and I need to calculate the response rate Pc based on the non-missing data in the control group, and for the control group, the results are based on Bernoulli (Pc) multiple filling. But I didn't find the relevant parameters in the PROC MI process at SAS. How do I do that?

Re: How to do multi-imputation based on Bernoulli method

JosvanderVelden — Mon, 13 Nov 2023 10:07:00 GMT

Maybe you can find some indications in this paper: https://support.sas.com/resources/papers/proceedings/proceedings/sugi30/113-30.pdf

Re: How to do multi-imputation based on Bernoulli method

SAS_Rob — Mon, 13 Nov 2023 14:20:00 GMT

This section in the documentation will be helpful in determining which method you should use based on the missing data pattern and the type of variable.

SAS Help Center: Imputation Methods

In your case you will likely want to decide whether to use either the DISCRIM or LOGISTIC method.