Blog Archives

Design top down, Code bottom up

May 22, 2017
By

Top-down design means designing from the client application programmer interface (API) down to the code. The API lays out a precise functional specification, which says what the code will do, not how it will do it. Coding bottom up means coding the lowest-level foundations first, testing them, then continuing to build. Sometimes this requires dropping […] The post Design top down, Code bottom up appeared first on Statistical Modeling, Causal…

Read more »

Fitting hierarchical GLMs in package X is like driving car Y

April 17, 2017
By

Given that Andrew started the Gremlin theme, I thought it would only be fitting to link to the following amusing blog post: Chris Brown: Choosing R packages for mixed effects modelling based on the car you drive (on the seascape models blog) It’s exactly what it says on the tin. I won’t spoil the punchline, […] The post Fitting hierarchical GLMs in package X is like driving car Y appeared…

Read more »

Bayesian Posteriors are Calibrated by Definition

April 12, 2017
By
Bayesian Posteriors are Calibrated by Definition

Time to get positive. I was asking Andrew whether it’s true that I have the right coverage in Bayesian posterior intervals if I generate the parameters from the prior and the data from the parameters. He replied that yes indeed that is true, and directed me to: Cook, S.R., Gelman, A. and Rubin, D.B. 2006. […] The post Bayesian Posteriors are Calibrated by Definition appeared first on Statistical Modeling, Causal…

Read more »

Ensemble Methods are Doomed to Fail in High Dimensions

March 15, 2017
By
Ensemble Methods are Doomed to Fail in High Dimensions

Ensemble methods By ensemble methods, I (Bob, not Andrew) mean approaches that scatter points in parameter space and then make moves by inteprolating or extrapolating among subsets of them. Two prominent examples are: Ter Braak’s differential evolution   Goodman and Weare’s walkers There are extensions and computer implementations of these algorithms. For example, the Python […] The post Ensemble Methods are Doomed to Fail in High Dimensions appeared first on…

Read more »

A fistful of Stan case studies: divergences and bias, identifying mixtures, and weakly informative priors

March 7, 2017
By

Following on from his talk at StanCon, Michael Betancourt just wrote three Stan case studies, all of which are must reads: Diagnosing Biased Inference with Divergences: This case study discusses the subtleties of accurate Markov chain Monte Carlo estimation and how divergences can be used to identify biased estimation in practice.   Identifying Bayesian Mixture […] The post A fistful of Stan case studies: divergences and bias, identifying mixtures, and…

Read more »

Stan Language Design History

February 28, 2017
By
Stan Language Design History

Andrew’s proposal At our last Stan meeting, Andrew proposed allowing priors to be defined for parameters near where they are declared, as in: parameters { real mu; mu ~ normal(0, 1); real sigma; sigma ~ lognormal(0, 1); ... I can see the pros and cons. The pro is that it’s easier to line things up […] The post Stan Language Design History appeared first on Statistical Modeling, Causal Inference, and…

Read more »

HMMs in Stan? Absolutely!

February 7, 2017
By
HMMs in Stan?  Absolutely!

I was having a conversation with Andrew that went like this yesterday: Andrew: Hey, someone’s giving a talk today on HMMs (that someone was Yang Chen, who was giving a talk based on her JASA paper Analyzing single-molecule protein transportation experiments via hierarchical hidden Markov models). Maybe we should add some specialized discrete modules to […] The post HMMs in Stan? Absolutely! appeared first on Statistical Modeling, Causal Inference, and…

Read more »

Stan JSS paper out: “Stan: A probabilistic programming language”

January 13, 2017
By
Stan JSS paper out:  “Stan: A probabilistic programming language”

As a surprise welcome to 2017, our paper on how the Stan language works along with an overview of how the MCMC and optimization algorithms work hit the stands this week. Bob Carpenter, Andrew Gelman, Matthew D. Hoffman, Daniel Lee, Ben Goodrich, Michael Betancourt, Marcus Brubaker, Jiqiang Guo, Peter Li, and Allen Riddell. 2017. Stan: […] The post Stan JSS paper out: “Stan: A probabilistic programming language” appeared first on…

Read more »

Stan 2.14 released for R and Python; fixes bug with sampler

January 2, 2017
By
Stan 2.14 released for R and Python; fixes bug with sampler

Stan 2.14 is out and it fixes the sampler bug in Stan versions 2.10 through 2.13. Critical update It’s critical to update to Stan 2.14. See: RStan 2.14.1 PyStan 2.14.0.0 CmdStan 2.14.0 The other interfaces will update when you udpate CmdStan. The process After Michael Betancourt diagnosed the bug, it didn’t take long for him […] The post Stan 2.14 released for R and Python; fixes bug with sampler appeared…

Read more »

How to include formulas (LaTeX) and code blocks in WordPress posts and replies

December 24, 2016
By
How to include formulas (LaTeX) and code blocks in WordPress posts and replies

It’s possible to include LaTeX formulas like . I entered it as $latex \int e^x \, \mathrm{d}x$. You can also generate code blocks like this for (n in 1:N) y[n] ~ normal(0, 1); The way to format them is to use <pre> to open the code block and </pre> to close it. You can create […] The post How to include formulas (LaTeX) and code blocks in WordPress posts and…

Read more »


Subscribe

Email:

  Subscribe