Ben Augustine
@bencaugustine.bsky.social
Ecological Statistics. I came for the charismatic megafauna and stayed for the problematic megamodels. Skeptic. github.com/benaug
Recently came across this Dorazio et al. paper showing that parameter expansion can make probit multispecies occupancy models much more efficient to sample.
besjournals.onlinelibrary.wiley.com/doi/full/10....
I wanted to learn a new trick so I made a nimble version.
github.com/benaug/Probi...
besjournals.onlinelibrary.wiley.com/doi/full/10....
I wanted to learn a new trick so I made a nimble version.
github.com/benaug/Probi...
November 19, 2025 at 6:23 PM
Recently came across this Dorazio et al. paper showing that parameter expansion can make probit multispecies occupancy models much more efficient to sample.
besjournals.onlinelibrary.wiley.com/doi/full/10....
I wanted to learn a new trick so I made a nimble version.
github.com/benaug/Probi...
besjournals.onlinelibrary.wiley.com/doi/full/10....
I wanted to learn a new trick so I made a nimble version.
github.com/benaug/Probi...
Lazy Coding Sunday
October 26, 2025 at 5:01 PM
Lazy Coding Sunday
After all the study planning, field work, model development/ refinement, and computation--Behold! the posterior distribution
October 15, 2025 at 3:05 PM
After all the study planning, field work, model development/ refinement, and computation--Behold! the posterior distribution
Matern Type III repulsive point processes in NIMBLE. Pretty cool approach and does not require a birth-death process for estimation, so it is fast and mixes well. Can add SCR observation model.
github.com/benaug/Mater...
github.com/benaug/Mater...
September 11, 2025 at 7:58 PM
Matern Type III repulsive point processes in NIMBLE. Pretty cool approach and does not require a birth-death process for estimation, so it is fast and mixes well. Can add SCR observation model.
github.com/benaug/Mater...
github.com/benaug/Mater...
Fun with multimodal posteriors
September 11, 2025 at 3:17 PM
Fun with multimodal posteriors
Simulate a point process, simulate uniform(0,1) ages for each point, put points in age order, then thin using a half-normal kernel.
August 26, 2025 at 1:12 PM
Simulate a point process, simulate uniform(0,1) ages for each point, put points in age order, then thin using a half-normal kernel.
Got generalized Matern Type III thinning process for point repulsion working using data augmentation. Cool approach, but pretty slow.
August 26, 2025 at 1:09 PM
Got generalized Matern Type III thinning process for point repulsion working using data augmentation. Cool approach, but pretty slow.
Figured out how this data augmentation approach to is related to that presented by Schofield and Barker (2014). Theirs can be specified with bugs code, which is very slick! As coded, mine runs 2x as fast with better ESS/iteration. Cool stuff. More formal details soon.
github.com/benaug/SCR-N...
github.com/benaug/SCR-N...
August 13, 2025 at 4:32 PM
Figured out how this data augmentation approach to is related to that presented by Schofield and Barker (2014). Theirs can be specified with bugs code, which is very slick! As coded, mine runs 2x as fast with better ESS/iteration. Cool stuff. More formal details soon.
github.com/benaug/SCR-N...
github.com/benaug/SCR-N...
Setting up final run for SMRing ground squirrels. Mb for marking process, negative binomial for sighting process, both sex-specific.
August 2, 2025 at 2:30 PM
Setting up final run for SMRing ground squirrels. Mb for marking process, negative binomial for sighting process, both sex-specific.
The overdispersion is strong in the SMR data set
July 12, 2025 at 12:46 PM
The overdispersion is strong in the SMR data set
Marginal version of SMR with interspersed marking and sighting and all sample types is pretty slick.
June 20, 2025 at 7:20 PM
Marginal version of SMR with interspersed marking and sighting and all sample types is pretty slick.
Top 5 things I want to write up from my Github repositories. Someone hire me for a while and we can pump these out. Will describe some more later.
May 20, 2025 at 5:07 PM
Top 5 things I want to write up from my Github repositories. Someone hire me for a while and we can pump these out. Will describe some more later.
Works with within home range resource selection, too. Here, availability distribution is BVN, use distribution is normalized product of availability distribution and RSF.
April 24, 2025 at 5:05 PM
Works with within home range resource selection, too. Here, availability distribution is BVN, use distribution is normalized product of availability distribution and RSF.
This looks promising approach for repulsion between individuals based on space use overlap instead of distance. Generalized Matern Type III Process thins a Poisson point process as a function of space use overlap, where space use is BVN.
April 24, 2025 at 4:58 PM
This looks promising approach for repulsion between individuals based on space use overlap instead of distance. Generalized Matern Type III Process thins a Poisson point process as a function of space use overlap, where space use is BVN.
Precognitive subskeeting? Damn, he's good.
April 21, 2025 at 9:48 PM
Precognitive subskeeting? Damn, he's good.
Pop simulator with habitat and density effects on reproduction, survival, and movement. Females only. Grey indicates 95% HR. Survival: red dots adult deaths, gold, 1 or more dependent deaths. Movement, red is adult HR relocation, gold subadult dispersal. Reproduction: gold indicates a new litter.
April 18, 2025 at 8:15 PM
Pop simulator with habitat and density effects on reproduction, survival, and movement. Females only. Grey indicates 95% HR. Survival: red dots adult deaths, gold, 1 or more dependent deaths. Movement, red is adult HR relocation, gold subadult dispersal. Reproduction: gold indicates a new litter.
The fastest way to Mb is to create first and subsequent capture histories and exposure matrices. Reduces data dimensions to equivalent of 2 occasions. Added this to my SCR Dcov repo.
github.com/benaug/SCR-D...
github.com/benaug/SCR-D...
April 9, 2025 at 6:57 PM
The fastest way to Mb is to create first and subsequent capture histories and exposure matrices. Reduces data dimensions to equivalent of 2 occasions. Added this to my SCR Dcov repo.
github.com/benaug/SCR-D...
github.com/benaug/SCR-D...
Last single-catch thing for now. Using Mb here, trap-happy. Obsmod parameters weakly identifiable esp. first cap p0 and sigma. What happens is trap happy inds almost always get to traps first and mostly block first capture events.
April 8, 2025 at 2:57 PM
Last single-catch thing for now. Using Mb here, trap-happy. Obsmod parameters weakly identifiable esp. first cap p0 and sigma. What happens is trap happy inds almost always get to traps first and mostly block first capture events.
The main problem with single catch traps is that you block more captures in high D areas than low D areas and underestimate Dcov betas using regular SCR. Can see that effect here.
April 7, 2025 at 3:13 PM
The main problem with single catch traps is that you block more captures in high D areas than low D areas and underestimate Dcov betas using regular SCR. Can see that effect here.
Latent ID SCR approach to Efford 2004 Oikos single catch traps modeling observed captures conditioned on the captures that would have happened if traps did not fill up or prevent captured individuals from being captured at other traps on the same occasion.
github.com/benaug/SCR_S...
github.com/benaug/SCR_S...
April 6, 2025 at 1:04 PM
Latent ID SCR approach to Efford 2004 Oikos single catch traps modeling observed captures conditioned on the captures that would have happened if traps did not fill up or prevent captured individuals from being captured at other traps on the same occasion.
github.com/benaug/SCR_S...
github.com/benaug/SCR_S...
This works for SCR book version
April 3, 2025 at 4:56 PM
This works for SCR book version
Pretty sure I solved the Efford 2004 version of single catch traps without knowing capture times or order. Framed as latent ID SCR model with a complicated thinning process. Looks spot on, will test formally.
Previously solved the version in Royle et al SCR book, but that is less realistic.
Previously solved the version in Royle et al SCR book, but that is less realistic.
April 3, 2025 at 4:53 PM
Pretty sure I solved the Efford 2004 version of single catch traps without knowing capture times or order. Framed as latent ID SCR model with a complicated thinning process. Looks spot on, will test formally.
Previously solved the version in Royle et al SCR book, but that is less realistic.
Previously solved the version in Royle et al SCR book, but that is less realistic.