I am analysing the interaction of determinants (i) with time (x) on the outcome (y) in longitudinal, time-variant data collected 1 to 8 times in around 700 participants. I want to discuss if the determinant (i) or the outcome (y) started changing first. If I understood correctly, I should use the previous values for the determinant (i) as lagged variable? However, I also received a recommendation to use the previous outcome variable (y) as lagged variable. Now I am confused.
Can anyone recommend any literature about assessing the temporality in longitudinal data in R?
Hi Anne-Marie,
I dunno about your divergent tips, but I know about wrangling time-series data in R:
The xts documentation may help you a bit.
https://cran.r-project.org/web/packages/xts/xts.pdf
https://ugoproto.github.io/ugo_r_doc/time_series_in_r_the_power_of_xts_and_zoo/
Of course Datacamp has some course on that too:
https://www.datacamp.com/cheat-sheet/xts-cheat-sheet-time-series-in-r
Please report back if you figured it out, I sense I'll need that too :)
All the best!
- Badges
- Science topic