Potential detrend correlations

classic Classic list List threaded Threaded
3 messages Options
SCO
Reply | Threaded
Open this post in threaded view
|

Potential detrend correlations

SCO
How does AstroImageJ handle detrend correlations?  For example, detrend parameter X-Width_T1 would likely be strongly correlated with detrend parameter Y-Width_T1.  Airmass would likely correlate (to a far lesser extent and less predictably) to FWHM_T1.  Other detrend parameters are completely independent of each other.  How extensive is the logic that AstroImageJ uses to assess such potential correlations and correct for each one appropriately?
Sycamore Canyon Observatory
https://scosci.com/
Reply | Threaded
Open this post in threaded view
|

Re: Potential detrend correlations

karenacollins
Administrator
The user may select any desired detrend parameter. If detrend optimization is used, it selects up to N justifiable detrend parameters based on the Bayesian information criterion, where N is set by the user in "Max Detrend Pars" and BIC threshold to justify an additional parameter is set in "Min BIC Thres", which should normally be set to "2". I wouldn't generally allow more than N=1 detrend parameters unless you have more baseline that the transit duration, and then maybe allow up to 2.
SCO
Reply | Threaded
Open this post in threaded view
|

Re: Potential detrend correlations

SCO
Sounds good.  Thanks!
Sycamore Canyon Observatory
https://scosci.com/