0
$\begingroup$

I have the following data/setting:

  • N persons assessed over time: different number of assessments; assessments at different time points
  • Variable Y: measured ~50 times (3 times per day)
  • Variable X: measured at 5 time points

It is possible to model Y ~ X? I was thinking in using mixed models and account for the two levels (assessments nested in individuals) but I don't know how to do it without losing information (as would happen if I use just 5 measures of Y).

In practice, I'm looking for two time series (for each individual) and I want to know if X affects the dynamic of Y over time.

Improve this question
$\endgroup$
2
  • $\begingroup$ You can always use the five x values with the corresponding five y values. Is interpolation of x values a reasonable option? $\endgroup$
    – Roland
    Commented Oct 1, 2021 at 6:32
  • $\begingroup$ @Roland, interpolation wouldn't be an option; use just 5 values of Y is possible but it means losing information. In practice, I'm looking for two time series (for each individual) and I want to know if X affects the dynamic of Y. I will edit my post to clarify these points. Thank you $\endgroup$
    – MDSF
    Commented Oct 1, 2021 at 9:57

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.