Quantcast
Channel: NITRC CONN : functional connectivity toolbox Forum: help
Viewing all articles
Browse latest Browse all 6859

Missing Covariates (use of NaNs), and what scan data should be included

$
0
0
Hello All
I am running a linear regression model where I have some missing covariates that are continuous (CSF_biomarker data). For those that are missing I have labeled them as NaNs (eg: 3 1 5 NaN 6 NaN) I was running my analysis as a linear regression with AllSubjects (so basically AllSubjects and CSF_bio, [0,1]).  However I was wondering what difference does it make to include AllSubjects or should I include only those in the analysis that I have CSF data on? 

For example should I create a new variable (Usable_Scans) and only highlight those scans that have available covariate data (CSF_biomarker data)? In other words my covariates would look like this:
Usable_Scans: 1 1 1 NaN 1 NaN
CSF_Bio:         3 1 5 NaN 6 NaN

and then run the data (Usable_Scans, CSF_Bio [0 1])

What are the differences here if any in the way the statistics are handled?  Thank you in advance for your help

Viewing all articles
Browse latest Browse all 6859

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>