turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- Rescale the data values for convergence

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

09-15-2015 12:04 PM

Hello SAS community,

I was reading one of the SAS global proceeding papers related to linear mixed models. That paper suggested to rescaling the data is one of the options to trouble shoot convergence issues in parameter estimation. I'm working on large dataset that consists of multple years and locations, and response variable is yield measured in tons/hectare. Since this dataset requires to estimate a large numbe of parameters, convergence is often a problem to estimate the parameters. I was thinking to rescale yield values to kgs or some other form.

Any suggestions or comments?

Thank you,

RVS

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

09-15-2015 12:28 PM

It might help to post a link to the paper so we can see which approach you may be contemplating.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to ballardw

09-21-2015 08:40 PM

Thanks for the response. Below is the link for the paper.

http://support.sas.com/resources/papers/proceedings12/332-2012.pdf

I'm refering to the Section II ("TROUBLE SHOOTING CONVERGENCE FAILURES IN MIXED MODELS") from the above paper.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

09-21-2015 08:53 PM

Tons per hectare should be a good scaling. Vaules are typically between 2 and 10, so there are not extremes. Converting to kg per hectare would probably cause more problems, considering the values would be ranging up 10000 or so. The variance-covariance matrix would have lots of very large numbers (for variances), and covariances which could be very small or very large.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

09-22-2015 08:32 AM

I suspect (unfortunately) that the model may be overspecified for the number of observations. Consequently, some random effect values are really sensitive to changes, and thus convergence becomes a problem.

My usual approach is to "sneak up on" the model I want to use, starting with something less complex, and working there until it converges and is meaningful in some sense. Then complexity can be added which, in mixed models, is almost always something in the random (or repeated, if in PROC MIXED) statement.

Steve Denham