Discover and read the best of Twitter Threads about #Stan

Most recents (6)

New paper for ‘ibis.iSDM’ now published in Ecological Informatics. Here I present a new R-package that allows the integration of different datatypes in SDMs. But it can do much more than that... #sdm #biodiversity #conservation
doi.org/10.1016/j.ecoi…
👇(1/8)
So one might ask why another R-package for correlative SDMs, and I generally agree. The specific role of the package for me and our group is to a) support all types of integration originally highlighted by Fletcher et al. and more (doi.org/10.1002/ecy.27…, @FletcherEcology) (2/8)
..., b) propose a modelling framework founded in Poisson Process Models and support Bayesian SDMs, c) enable easy construction of scenarios and projections and d) have a modular coding framework that can be easily updated with further functionalities (3/8)
Read 8 tweets
1/ Bayesian inference is a powerful statistical framework that allows us to estimate the probability distribution of parameters based on data and prior knowledge. And R has a variety of packages for implementing Bayesian analysis! #rstats #datascience #Bayesian
2/ First up, there's 'rstan', which provides R interfaces to the popular Stan modeling language. It includes a suite of powerful algorithms for Bayesian inference, including Markov Chain Monte Carlo (MCMC) sampling and Variational Bayes (VB). #rstats #stan
3/ Another popular R package for Bayesian analysis is 'JAGS', which stands for Just Another Gibbs Sampler. It's an alternative to Stan that uses a different MCMC algorithm, and can be easier to use for some types of models. #rstats #JAGS
Read 7 tweets
1) Wherever I look these days I see #occult symbols and imagery. Was in Hyde Park yesterday and saw this bus with a big ad for #WitchesMovie. Image
2) There was this ad for #HamiltonMusical nearby in Elizabeth Street. Not about witchcraft, sure. But it still featured the star motif. This is basically an inverted pentagram. #symbolism #Sydney Image
3) This was display in boutique on Elizabeth Street. Has a #Christmas theme and it's not surprising that star (of Bethlehem) symbolism would be part of that. But what I think is interesting is that the stars are more prominent than the tree. Image
Read 33 tweets
New paper: Tooling-up for infectious disease transmission modelling. The Intro to the #Epidemics special issue on Computational Methods for Modelling an Infectious Disease (COMMAND) doi-org.ez.lshtm.ac.uk/10.1016/j.epid… @GrahamMedley @EleanorMRees @N_R_Waterlow @cmmid_lshtm In the series:
1. Choices and trade-offs in inference with infectious disease models by @sbfnk and A King doi.org/10.1016/j.epid… #IDmodelling #modelchoice
2. Desirable BUGS in models of infectious diseases by @kathmoreilly @drrachellowe and others doi.org/10.1016/j.epid… #JAGS #graphs
Read 8 tweets
How do you test 941 voxels without having to correct for multiple comparisons? 🔥🔥
Build 1⃣ Bayesian multilevel model and estimate 1⃣ multidimensional posterior distribution. No correction needed! (see paper below)🧵 Posterior probability that ...
Inspired by work with @gangchen6 on Bayesian multilevel models at the ROI level:
link.springer.com/article/10.100…
🧵
Lab effort to apply the framework at the ROI and in a voxelwise manner (just insula for now) to study the effect of controllability on shock-plus-sound stressors:
biorxiv.org/content/10.110…
Looking forward to running whole-brain soon.🧵
Read 5 tweets
1/7 In 2017, Hedge, Powell, and Sumner showed that robust cognitive tasks are unreliable, which calls into question the use of behavioral tasks for studying individual differences. In this blog post, I show that this conclusion is misguided (haines-lab.com/post/thinking-…)
2/7 Specifically, Hedge et al. found that robust effects such as the Stroop effect have test-retest correlations in the range of .5 to .6 (r = .5 in the plot of their Stroop effects shown here), which severely impacts our ability to rank the performance of individual subjects.
3/7 However, this conclusion is based on a model of behavior does not account for variability at the individual-subject level. While this may not seem very problematic at first glance, the assumption of no measurement error leads to substantially biased inference.
Read 7 tweets

Related hashtags

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!