Programming the statistical procedures from SAS

Two part model for health care costs

Accepted Solution Solved
Reply
New Contributor
Posts: 2
Accepted Solution

Two part model for health care costs

Can anyone provide or direct me to SAS code for two part models in GLM.  I am analyzing cost differences between patients who have experienced a poisoning and those who have not.  There are lots of zeroes in the data and it is skewed. From what I've read, the first part of the model would be a logistic regression and the second would be a regression with a gamma distribution and log link.  I am especially interested in how to combine the results of the two regressions to develop estimates of cost differences between the two groups.


Accepted Solutions
Solution
‎07-07-2017 08:22 AM
SAS Employee
Posts: 29

Re: Two part model for health care costs

Try using PROC FMM:

 

Your SAS code should look something like this:

 

proc fmm data=rowdata;

  model response = age income avgexp / dist=binary;

   model response = age income avgexp / dist=gamma;

   model response = / dist=constant; 

run;

 

The last MODEL statement specifies a constant distribution with all mass at zero for the zero target group.

 

To understand PROC FMM and finite mixture models take a look at:

Funda

 

 

 

View solution in original post


All Replies
Super User
Posts: 18,549

Re: Two part model for health care costs

To get you started see the papers at Lexjansen.com

 

http://lexjansen.com/search/searchresults.php?q=two%20stage%20model

 

FYI - This is a good place to start research on any SAS topic Smiley Happy

Solution
‎07-07-2017 08:22 AM
SAS Employee
Posts: 29

Re: Two part model for health care costs

Try using PROC FMM:

 

Your SAS code should look something like this:

 

proc fmm data=rowdata;

  model response = age income avgexp / dist=binary;

   model response = age income avgexp / dist=gamma;

   model response = / dist=constant; 

run;

 

The last MODEL statement specifies a constant distribution with all mass at zero for the zero target group.

 

To understand PROC FMM and finite mixture models take a look at:

Funda

 

 

 

New Contributor
Posts: 2

Re: Two part model for health care costs

nvcarroll54, were you able to find the code for this?

 

Thanks

New Contributor
Posts: 2

Re: Two part model for health care costs

Nothing simple enough for me to understand and use

##- Please type your reply above this line. Simple formatting, no
attachments. -##
New Contributor
Posts: 2

Re: Two part model for health care costs

I know what you mean... i've been trying to figure this out for over a month now.

 

Turns out, it's way easier to do it in Stata, but my dataset is way to large for Stata to handle, so I've had to resort back to SAS.

 

If your dataset is not very large, try running this in Stata. You have to install command twopm, and it's literally just few lines of code. 

 

Goodluck,

 

☑ This topic is solved.

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

Discussion stats
  • 5 replies
  • 943 views
  • 5 likes
  • 4 in conversation