Building models with SAS Enterprise Miner, SAS Factory Miner, SAS Visual Data Mining and Machine Learning or just with programming

Imputation for categorical variables

Reply
Contributor
Posts: 37

Imputation for categorical variables

Hello Everyone,

I was just wondering what the best imputation method for categorical variables is. Considering many categorical variables (with exception of variables such as postcodes etc) dont have too many levels, will it not be misleading to impute their missing values?

 

I have always thought creating a 'one-size-fits-all' variable such as CCC and imputing that for every missing level might be more reasonable. Can someone help out with any ideas or suggestions?

Thanks

Trusted Advisor
Posts: 1,909

Re: Imputation for categorical variables

There's lots of discussion of this subject if you go to your favorite search engine and type in

 

imputing categorical variables

Super User
Posts: 19,768

Re: Imputation for categorical variables

That's why you do multiple imputation to reduce chance of bias AND to understand the effect of the imputation. 

 

This isn't a SAS question though, it's a statistics question better answered by statisticians or Google. 

Ask a Question
Discussion stats
  • 2 replies
  • 132 views
  • 0 likes
  • 3 in conversation