Hi,
Answers inline with questions below
> On 15 Aug 2022, at 19:24, Monica Keith <[log in to unmask]> wrote:
>
> Hi, I have two follow up questions to this post:
>
> 1. Is it wrong to first convert the values of the covariate (i.e. age) to z-scores, and then demean those values? Long time ago I was recommended to do that but now I don’t know if that’s the right thing to do. How does it affect the model to first convert the values to z-scores?
Assuming you have a known distribution of ages you _could_ do this, but most models will be fine with plain de-demeaned age. The effect on the model is to change the relationship with age - rather than testing for a linear relationship with age, you are now testing for a linear relationship with standard-deviations of age. Without knowing the ground-truth relationship between your data and age, it’s impossible to canonically say which is a “truer” model...
>
> 2. The -D flag, should it be ONLY used when NOT including the column of ones (the constant EV) or can it be used always, even when the constant EV is included in the model?
Correct - demeaning the regressor(s) representing the global mean will potentially result in issues when estimating model parameters..
>
> Thanks!
Kind Regards
Matthew
--------------------------------
Dr Matthew Webster
FMRIB Centre
John Radcliffe Hospital
University of Oxford
>
> ########################################################################
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
>
> This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|