Imputation Techniques for missing values in SAS

Question — Thu, 14 Jan 2021 16:33:59 GMT

Hi,

I was wondering if you would be able to help regarding imputation techniques in SAS.
Basically I am working on a project were we have to match survey data with other data on a match key.

The survey has around 33,000 people and we get a 95% match but we need to impute the 5 % missing. But in the 95% matched we are only interested in around 200 people who have a certain product( as an example, a home Insurance). But in total we need to have 500 people with a home insurance.

The objective is to find lookalike of the 200 (Home insurance customers) matched in the 5% unmatched so we get to 500 . Some have suggested hot decking technique but I would like to know if there is a better technique for this (in SAS).

Not sure if it makes sense , if not please let me know and I will clarify

Your help would be much appreciated.

Thank you

Re: Imputation Techniques for missing values in SAS

Watts — Thu, 14 Jan 2021 17:54:22 GMT

PROC SURVEYIMPUTE provides several imputation methods for handling missing values in survey data -- including some traditional hot-deck imputation methods and a fractional hot-deck method.

Re: Imputation Techniques for missing values in SAS

Question — Thu, 14 Jan 2021 18:38:32 GMT

Great! Thank you ...I will try it and see how it goes 🙂

Best Wishes

topic Re: Imputation Techniques for missing values in SAS in SAS Data Science

Imputation Techniques for missing values in SAS

Re: Imputation Techniques for missing values in SAS

Re: Imputation Techniques for missing values in SAS