Great Horned Owl “Hooting Season” Over

While I still hold out some hope, we’re largely past the key window for detecting great horned owls vocalizing.

“In Wisconsin, territorial hooting ends in mid-Feb, in keeping with timing of first eggs laid…” – Birds of North America

Across the state, birds are probably on eggs by now and thus less conspicuous. Since this might be my last winter here, I may not find this species breeding in my atlas block, without some serious luck and/or spotting a nest!

Fastest Fitting for Bayesian Models?

Right now, I’m focused on computational efficiency (and thereby speed) of fitting Bayesian models. For my dissertation I used R-INLA, so in this “lab notebook” of a blog I need to jot down my thoughts around some current methodological hangups. Right now, even though it seems to be the fastest thing available, it’s honestly still not quite fast or  efficient enough for the full analysis I want to run. It’s taking too long for a re-analysis. I’m exploring whether or not it’s still the best thing available.

  • R-INLA
    • pros…
      • fast, as compared to other fitting algorithms
    • cons…
      • parallel processing (considering how much processing is sometimes required) can easily hog shared server resources: a postdoc I worked with and I talked about this a fair amount
      • crashes without clear indication of what happened: it’s a bit of a black box
      • can’t always estimate some parameters of interest
  • Hamiltonian Monte Carlo (Nov. 2016)
    • Stan: faster than BUGS (MCMC)
    • No-U-Turn Sampler
    • faster than INLA?
  • Bayesian variable selection regression (2017)

Valentine’s Evening Luck: Barred Owl!

I was so thrilled to have my 2nd identifiable sighting of an owl that might breed here last night: I saw my 1st ever barred owl in my atlas block! Unfortunately, it was just hunting, so I didn’t have any breeding activity clues. It caught my eye when it flew up to a telephone line, and then hunted the ditch below. It dove, but I’m not sure if it caught prey. Then, it flew off from the ground toward the lake. I lost it as it flew through a yard behind a house.

Here in MN they should start ramping up vocalizations soon! I wonder where it’s headed, and will have to go to potential areas nearby in the evenings to listen.

Considering Planning a “Spring Break” Trip

So I don’t really get spring break anymore, but I’m thinking about heading to the Black Hills Mar. 24-Apr. 1 this year. I want to bee-line to the Badlands (get to Blue Earth, MN Friday night and drive the rest of the way Saturday). I hope to catch some straggling gray-crowned rosy-finches before they’ve left for their breeding grounds. Canyon wrens should be singing by then. Maybe sage thrashers will moving through? I wonder if it’s too early for Cordilleran flycatcher.

The Latest Bayes in R

What good fortune that I looked up advances in Bayesian methods the day a new task view was published! If you’re just starting with Bayes, they even have packages listed here designed to help you learn. I’ve been using R-INLA and for a few reasons I became curious what else was out there. I’m still assuming Laplace estimations are faster than Markov chain Monte Carlo, so there are more on the list that might be good to check out if I abandon that method altogether.

  • LaPlace’s Demon, iter-lap: This one appealed to me as an alternative to R-INLA
  • arm, RSGHB: might be worth checking out for hierarchical models
  • Bayes images: image analysis
  • Bayes meta, bspmma: meta-analysis
  • Bayes tree,tgp: Bayesian additive regression trees (BART)
  • Bayes QR: quantile regression
  • Bayes var-sel: variable selection
  • deal, ebdb Net, grain, network change (etc. see this link): network analysis
  • FME: an alternative to deSolve
  • hbsae: small area estimates
  • spBayes: spatial
  • spikeSlabGAM: geo-additive
  • spTimer: space-time
  • ramps: geo-statistical

There are also interesting choices for model averaging (and ensembles), time series, extreme value, clustering, threshold, change point and auto-regressive models.