How sas handle missing data in the regression

Accepted Solution Solved
Reply
Frequent Contributor
Posts: 122
Accepted Solution

How sas handle missing data in the regression

Hi,

I want to do a regression. There are four independent variables x1,x2,x3,x4, and dependent variable y. There are some missing values in the independent variables. See below for an example.

y     x1     x2     x3     x4

xx     xx     xx     ..     ..

xx     xx     xx     ..     ..

xx     xx     xx     xx     xx

xx     xx     xx     xx     xx

...

xx means having data. ".." means missing data. If I run proc reg to regress y on x1, x2, x3, x4, How SAS will deal with this case?

Thanks.


Accepted Solutions
Solution
‎07-16-2013 08:11 AM
Super Contributor
Posts: 644

Re: How sas handle missing data in the regression

Rows with any missing values will be ignored.  That is why you might want to do imputation.

Richard

View solution in original post


All Replies
Contributor
Posts: 26

Re: How sas handle missing data in the regression

Various methods used to do missing value imputation

1.) Replacing missing values with mean, minimum or maximum values

 

2.) Replacing missing values with the mean, minimum or maximum of the most frequently appearing values.

 

3.) Replacing missing values by carrying forward or backward, or by averaging the values adjacent to missing data.

 

4.) Random generation of the missing value based on a sample mean.

Frequent Contributor
Posts: 122

Re: How sas handle missing data in the regression

Posted in reply to supersonic

Thank you for your answer. I want to know if I don't do anything about this. How SAS will handle with missing data?

Solution
‎07-16-2013 08:11 AM
Super Contributor
Posts: 644

Re: How sas handle missing data in the regression

Rows with any missing values will be ignored.  That is why you might want to do imputation.

Richard

🔒 This topic is solved and locked.

Need further help from the community? Please ask a new question.

Discussion stats
  • 3 replies
  • 200 views
  • 0 likes
  • 3 in conversation