Matthijs Hollanders

Anyone ever use the geometric Poisson distribution? If I've implemented it correctly in Stan, it's giving me better fits than negative binomial according to PSIS-LOO-CV with a big count occupancy model.

occARU now supports multiseason occupancy. The package (1) detects which sites were sampled in each season and then (2) fits a continuous time dynamic occupancy model for repeatedly sampled sites. 1/n mhollanders.github.io/occARU/artic...

The ecological parameters are then recovered as: logit(psi) = alpha[1] + lambda[1] * eta log(gamma) = alpha[2] + lambda[2] * eta log(epsilon) = alpha[3] + lambda[3] * eta Maybe not everyone's cup of tea, but drastically reduces model complexity; nice for multispecies models! 5/n

occARU now supports multiseason occupancy. The package (1) detects which sites were sampled in each season and then (2) fits a continuous time dynamic occupancy model for repeatedly sampled sites. 1/n mhollanders.github.io/occARU/artic...

For the stats-nerds, you can actually analytically compute the log of the matrix exponential for a 2x2 generator matrix. See log_matrix_exp() here. 3/n github.com/mhollanders/...

occARU now supports multiseason occupancy. The package (1) detects which sites were sampled in each season and then (2) fits a continuous time dynamic occupancy model for repeatedly sampled sites. 1/n mhollanders.github.io/occARU/artic...

occARU now supports sites within regions, e.g., multiple arrays of camera traps or microphones. Random (zero-sum) region-level intercepts are estimated for both occupancy and detection, and spatial site effects are computed per-region, which is much faster! 1/2 mhollanders.github.io/occARU/

occARU: Bayesian (multispecies) occupancy models for ARU data using @mc-stan.org. Automated recording units (ARUs) like camera traps produce rich time series which warrant going beyond occupancy and focusing on detection rates. 1/n mhollanders.github.io/occARU/

occARU accommodates site-level predictors for occupancy and detection and additionally site-by-survey level predictors for detection rates. Continuous, categorical, and ordinal predictors are currently supported. Model complexity is handled via global-local shrinkage priors. 4/n

occARU: Bayesian (multispecies) occupancy models for ARU data using @mc-stan.org. Automated recording units (ARUs) like camera traps produce rich time series which warrant going beyond occupancy and focusing on detection rates. 1/n mhollanders.github.io/occARU/

occARU uses hierarchical multispecies Gaussian processes to infer spatial and temporal variation in detection rates, and estimates interspecific correlations in these processes. These effects are orthogonally projected to retain inference on fixed effects. 2/n

New pre-print! I cover a range of open capture-recapture models (single survey/robust design, multistate/multievent, in (Cormack-)Jolly-Seber variants) in Stan, and provide efficient log likelihood functions. I also introduce a method to account for unequal survey intervals in the entry process.

In (Gaussian) linear models, we usually put normal priors for regression coefficients. How do we feel about putting gamma(a, a) priors in something like Poisson regression? Using rates for rates seems nice; the geometric means are still 1, meaning the priors are centered on no multiplicative effects

Hey #stats people, what do we think about interpreting coefficients in cloglog binomial regressions? Since everyone hates odds ratios I wonder if interpreting hazard ratios is easier.

Why do we peer review for free?