search

nespinoza / juliet

A versatile modelling tool for transiting and non-transiting (single and multiple) exoplanetary systems
MIT License
55 stars 30 forks source link

Enhancement suggestion: Simultanous GP fitting (shared hyperparameters) for activity indicators and RV/Transit data #35

Open StephanStock opened 4 years ago

StephanStock commented 4 years ago

Hi Nestor,

One thing I would love juliet to be capable, is the combined GP modelling of activity indicator/photometry with RV or Transit data. In principle it is very similar to sharing hyperparameters between different instruments and therefore maybe its not to hard to implement. One must only exclude the Keplerian model to be applied on the activity indicators or photometry.

I think this is a good way to constrain your GP and I see more and more studies which use this technique.

Best, Stephan

PS: I might be influenced a lot by the Sagan workshop this week.

nespinoza commented 4 years ago

Hi @StephanStock,

Yeah, this has been in the backburner for me for a while. I think it would mean adding an extra "dataset" type to juliet, so as we have "photometry" and "RVs" we would also have "activity indicators" that as you say would be very similar to the RV models, only ommiting the Keplerians.

I'll leave it in the backburner for now, hoping I have time to implement this in the near future. It should take me about a couple of days, but those are days I currently don't have sadly.

N.