| |
|
|
| |
EVENT DETAILS |
<P>Learn Bayesian Data Analysis (BDA) & Markov chain Monte Carlo (MCMC)computation using Stan with Bayes Master<STRONG><A HREF="http://andrewgelman.com/" TARGET="_blank" REL="noopener noreferrer noopener nofollow noopener noreferrer nofollow nofollow noreferrer nofollow">Andrew Gelman</A></STRONG>and Stan developers<A HREF="http://jgabry.github.io/about/" TARGET="_blank" REL="noopener noreferrer noopener nofollow noopener noreferrer nofollow nofollow noreferrer nofollow"><STRONG>Jonah Gabry</STRONG></A>and <STRONG><A HREF="https://betanalpha.github.io/" TARGET="_blank" REL="noopener noreferrer noopener nofollow noopener noreferrer nofollow nofollow noreferrer nofollow">Michael Betancourt</A></STRONG>.</P>
<P>This<STRONG>three-day</STRONG>course consists of three main themes: Bayesian inference & computation; the Stan programming language; applied statistics.</P>
<P>Participants will receive a copy of Andrew Gelman's landmark book<STRONG><A HREF="http://www.stat.columbia.edu/~gelman/book/" TARGET="_blank" REL="noopener noreferrer noopener nofollow noopener noreferrer nofollow nofollow noreferrer nofollow">Bayesian Data Analysis</A></STRONG>.Proceeds from the class support<STRONG>further development</STRONG>of<A HREF="http://mc-stan.org/" TARGET="_blank" REL="noopener noreferrer noopener nofollow noopener noreferrer nofollow nofollow noreferrer nofollow">Stan</A>and the <A HREF="http://www.nyhackr.org" TARGET="_blank" REL="noopener noreferrer noopener nofollow noopener noreferrer nofollow nofollow noreferrer nofollow">New York Open Statistical Programming Meetup</A>.</P>
<P>Before class everyone<STRONG>should install</STRONG><A HREF="http://www.r-project.org/" TARGET="_blank" REL="noopener noreferrer noopener nofollow noopener noreferrer nofollow nofollow noreferrer nofollow">R</A>,<A HREF="http://www.rstudio.com/" TARGET="_blank" REL="noopener noreferrer noopener nofollow noopener noreferrer nofollow nofollow noreferrer nofollow">RStudio</A>and<A HREF="http://mc-stan.org/rstan.html" TARGET="_blank" REL="noopener noreferrer noopener nofollow noopener noreferrer nofollow nofollow noreferrer nofollow">RStan</A>on their computers. If problems occur please join the<A HREF="https://groups.google.com/forum/#!forum/stan-users" TARGET="_blank" REL="noopener noreferrer noopener nofollow noopener noreferrer nofollow nofollow noreferrer nofollow">stan-users</A>group & post any questions. It is<STRONG>important</STRONG>that all participants get Stan running & bring their laptops to the course.</P>
<P>Class structure & example topics for the three days:</P>
<H2>Day 1: <STRONG>Foundations</STRONG></H2>
<UL>
<LI>Foundations of Bayesian inference</LI>
<LI>Foundations of Bayesian computation with Markov Chain Monte Carlo</LI>
<LI>Intro to Stan with hands-on exercises</LI>
<LI>Real-life Stan</LI>
<LI>Bayesian Workflow</LI>
</UL>
<H2>Day 2: <STRONG>Linear & Generalized Linear Models</STRONG></H2>
<UL>
<LI>Foundations of Bayesian regression</LI>
<LI>Fitting GLMs in Stan (e.g., logistic regression, Poisson regression)</LI>
<LI>Diagnosing model misfit using graphical posterior predictive checks</LI>
<LI>Little data: How traditional statistical ideas remain relevant in a big data world</LI>
<LI>Generalizing from sample to population (surveys, xbox example, etc)</LI>
</UL>
<H2>Day 3:<STRONG>Hierarchical Models</STRONG></H2>
<UL>
<LI>Foundations of Bayesian hierarchical/multilevel models</LI>
<LI>Accurately fitting hierarchical models in Stan</LI>
<LI>Why we dont (usually) have to worry about multiple comparisons</LI>
<LI>Hierarchical modeling & prior information</LI>
</UL>
<H2>Topics</H2>
<P>Specific topics on Bayesian inference & computation include, but are not limited to:<BR></P>
<UL>
<LI>Bayesian inference & prediction</LI>
<LI>Naive Bayes, supervised, & unsupervised classification</LI>
<LI>Overview of Monte Carlo methods</LI>
<LI>Convergence & effective sample size</LI>
<LI>Hamiltonian Monte Carlo & the no-U-turn sampler</LI>
<LI>Continuous & discrete-data regression models</LI>
<LI>Mixture models</LI>
<LI>Measurement-error & item-response models</LI>
<LI>Specific topics on Stan include, but are not limited to:</LI>
<LI>Reproducible research</LI>
<LI>Probabilistic programming</LI>
<LI>Stan syntax & programming</LI>
<LI>Optimization</LI>
<LI>Warmup, adaptation, & convergence</LI>
<LI>Identifiability & problematic posteriors</LI>
<LI>Handling missing data</LI>
<LI>Ragged & sparse data structures</LI>
<LI>Gaussian processes</LI>
</UL>
<P>For more information please <A HREF="https://www.landeranalytics.com/contact" TARGET="_blank" REL="noopener noreferrer noopener nofollow noopener noreferrer nofollow nofollow noreferrer nofollow">contact us</A>.</P>
<P><BR></P>
|
|
|
|
|
|
