Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

Using time serie to predict another variable in JASP

Hi. I'm new to this kind of analysis, but I would love to make it work in JASP.

I want to analyze head rotation (3 axes eulerangles rotation, x y z in absolute) in space (15hz) during 10-13 minutes. I have 35 participants with approximatively 10 000 measures per axis per participants.

For each participant I measured another variable called A (continuous from 0 to 15, but 1 measure by participant).

I have been doing some traditionnal analysis (coefficient of variation, speed etc), but I would like to train a ML model to explore the possibility of predicting A from the time series (even if its only on 1 axis). Is there any way to do it in JASP ? Thanks in advance !

Comments

  • Hmm. We do have a range of ML techniques, and we have the start of some time series tools (although much of this is under development), but I don't think we have what you are looking for at the moment. I would encourage you to make a GitHub request (for details see https://jasp-stats.org/2018/03/29/request-feature-report-bug-jasp/), preferably together with the suggestion for an R package to implement.

    Cheers,

    E.J.

Sign In or Register to comment.