Conservation science is interdisciplinary & collaborative. A great part is stepping back & admiring the sheer talent of your coauthors. Fun to watch Drs Nelson, Ward, & Linden turn our chats into Bayesian code. Statistical alchemy!
https://t.co/YJbTUK9JGD
Chuffed to have this out. Really hope its taken in the spirit meant:
a) to put in black and white some advice we find ourselves giving an awful lot (often in response to reviewers)
b) a complement to a paper we regularly recommend folks read -> @Todd_W_Arnold 2010 (1/2)
Looking for ways to evaluate model fit when conducting step-selection analyses or when fitting hidden Markov models? Check out our new preprint! With @signer_j and Smith Freeman #rstats
@vdVeenB @chrissuthy@RSocPublishing The phrase “model selection” often colloquially refers specifically to variable selection, even though the former can involve a larger set of decisions. Maybe this is a shortcoming of the discourse regarding model selection in the ecological literature…
I often get asked about Bayes factors and confidence intervals as alternative to p-values. I don’t think either of these is any solution at all. https://t.co/VqosMxVAbM
A bonus @CREEM_cake seminar today by Dan Linden (@dwlinden), research statistician at @NOAA.
Discussing discrete vs. continuous models for right whale population dynamics, and enjoying some #cake obviously!
@vdVeenB To be clear, the measurement model is making predictions for the state model. The latter is where inference is desired, so it still happens as a result.
@vdVeenB The original motivation for B&A was hierarchical models where the measurement error (e.g., capture probability) has a potentially complex structure. Since the purpose is making predictions for the state variable, this seems like the perfect application of model selection, no?
Our session on population dynamics and hierarchical modelling features whales, birds, and fish! Don't miss out on these exciting talks! Remember that #ISEC2022 is a hybrid event so there is still time to register as a virtual attendee!
@SEEC_UCT @AlisonKetz @hanna_mccaslin
@oaggimenez@Fuller_Lab@andyroyle_pwrc thanks @oaggimenez! I kinda wish we had fleshed that intro into a full blown conceptual review, as others eventually did. That Efford & Dawson (2012) paper was underrated...
Only 1 mo. to apply to learn #cameratrapping analysis and study design from the experts [Jim Nichols, Jim Hines, @andyroyle_pwrc, @dwlinden, @chrissuthy]. Time to dig into that unspent travel money and build out your camtrap toolkit. June 6-17. #Rstats
https://t.co/JSBWlE5OzE
📣Postdoc opportunity in quantitative ecology, supporting @NOAAFish_NEFSC with population assessment of North Atlantic right whales. Novel statistical models and a great team of scientists across multiple federal agencies and NGOs!
https://t.co/9TFahyQ2uJ