search

BrunaLab / HeliconiaDemography

Effects of drought and habitat fragmentation on Heliconia acuminata.
Other
0 stars 1 forks source link

Correct CIs for smooths? #93

Closed Aariq closed 3 years ago

Aariq commented 3 years ago

Be certain that the correct combination of overall_uncertainty and unconditional are chosen in smooth_estimates(). Report choices and reference Marra and Wood 2006. Figuring it out here: https://github.com/gavinsimpson/gratia/discussions/110

Aariq commented 3 years ago

Gavin Simpson suggests simultaneous intervals, but it seems like the confint.gam() method in gratia is computationally intensive and may not work with dlnm. The Bayesian confidence intervals that I'm using have good across-function coverage though, especially with unconditional=TRUE. (see section 4.3 here: https://www.frontiersin.org/articles/10.3389/fevo.2018.00149/full)