[Homer-users] DriftOrder correct

[Boas, David]

We typically use 3.

What do you mean your r values continue to increase? What are you correlating?

From: <homer-users-bounces at nmr.mgh.harvard.edu> on behalf of Jonathan Perry <walwynperry at gmail.com>
Reply-To: "homer-users at nmr.mgh.harvard.edu" <homer-users at nmr.mgh.harvard.edu>
Date: Tuesday, November 21, 2017 at 6:58 PM
To: "homer-users at nmr.mgh.harvard.edu" <homer-users at nmr.mgh.harvard.edu>
Subject: [Homer-users] DriftOrder correct

Hello all,

Does anyone have an idea of how much driftOrderCorrection is too much?  I am using the multiple gaussian idxBasis in the GLM analysis in Homer.  I have noticed that I can increase my driftOrderCorrection up to an order of 10+ and my r values continue to increase.  I have a sneaking suspicion this is either cheating, or this is excessive and will lead to overfitting or something similar.

Does anyone know any information for this?  I have been looking with little to no success.

Thank you all,


Jonathan Perry

Master's Student

Department of Engineering Technology

University of Houston

 4734 Calhoun Rd., Room# 123

Houston, TX 77204-4021

Tel: (<tel:%28713%29%20743-4159>512)925-4910
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/homer-users/attachments/20171122/9f2c936e/attachment.html