Bayesian Modeling with R and Stan (Reupload)
Ғылым және технология
Recent advances in Markov Chain Monte Carlo (MCMC) simulation have led to the development of a high-level probability modeling language called Stan. In this presentation, Sean Raleigh will give a gentle introduction to Bayesian inference using R and Stan.
Sean Raleigh received his Ph.D. in mathematics from U.C. San Diego, specializing in geometric topology and knot theory. He is a professor of mathematics at Westminster College and currently chairs the data science program. As part of Sean's professional work, he advocates for Bayesian methods in data analysis and co-directs QUARC, the Quantitative Analysis and Research Cooperative.
Пікірлер: 23
I wish our econometrics professors were teaching like you. Outstanding!!
Just learned more in this brilliant lecture than the whole semester!! GRACIAS!!! ❤️🇲🇽❤️
One of the bast classes ever! Congrats.
oh Waaw! I paid around a 10 dollars to learn this in my university and this guy does it better than anyone is had seen!
This was incredibly helpful, thanks for sharing this video!
wow!! could be the best lecture i ever heard! thank you!!
Really helpful thank you very much!!
Great lecture, thanks!
Fantastic lecture. Thank you
I know the beauty of Bayesian now!
Great lecture!
brilliant man. Period.
Running bin_unif
wow this resolves my confusion between the 2
Can you please clarify the fact the no multiple testing correction is needed for Bayesian approach - thank you
hello, how can i define a prior from a previous experiment?
Can you please provide the link to download the dataset used in the video?
@SeanRaleigh
2 жыл бұрын
Unfortunately, the data is confidential. So while I can share the analysis, I'm not allowed to share the raw data. Sorry about that.
You should only use wide priors if you have no information about your data. It should usually be possible to come up with more informative prior distributions by reading other studies in your area or talking to experts.
@SeanRaleigh
2 жыл бұрын
Yes and no. With quality data, the prior is overwhelmed by the data, so the prior shouldn't matter too much as long as it's in the right ballpark. And translating abstract information from other studies and experts into a prior distribution is surprisingly challenging.
Wow