Interpolation options and iglevel options

unread,

Dec 13, 2021, 12:59:33 AM12/13/21

to R package GGIR

Hi all,

We are going to use GGIR to analyse the UKbiobank dataset and would like your opinion on the differences between the parameter options.

For iglevels we have simply used the default "iglevels=c(seq(0,4000,by=25),8000)", is this the convention?

For the interpolation option, we do not know when to use "1=linear" or "2=nearest neighbour" and their pros & cons.

Lastly, if you could provide any comments in general on why the defaults in GGIR are chosen as they are, that would be very helpful!

Warm regards,

Po

unread,

Dec 13, 2021, 10:55:45 AM12/13/21

to R package GGIR

Hi,

I recently read this paper using both interpolation options, might be helpful: https://journals.humankinetics.com/view/journals/jmpb/4/4/article-p298.xml

Best,
Jairo

unread,

Dec 13, 2021, 6:16:21 PM12/13/21

to R package GGIR

Hi Jairo,

Thank you for this resource, much appreciated.

Warm regards,

Po

Reply all

Reply to author

Forward