Hi all, This post deals with a strange phenomenon in R that I have noticed while working on unbiased MCMC. Reducing the problem to a simple form, consider the following code, which iteratively samples a vector ‘x’ and stores it in a row of a large matrix called ‘chain’ (I’ve kept the MCMC […]
I blogged earlier on a problem with Bayesian model averaging (BMA) and gave some links to new work that chips away at it. The interesting thing about that new work is that it stays very close to traditional BMA while acknowledging that all models are m...
My reply: Thanks for sending. I can’t be sure about everything they’re doing but the paper looks reasonable to me. I expect there are various ways that the analysis could be improved, but on a quick look I don’t see anything obviously wrong with it, and the authors seem to know what they are doing. […] The post A reporter…
StanCon2018 is shaping up nicely as a unique opportunity to immerse oneself in all things Stan, meet Stan developers and fellow users. Registration is still open, but spots are filling up fast. We’re at 130 registrants and counting! The draft schedule is now up. We have 16 accepted talks and 6 invited talks. Posters are […] The post StanCon2018: one…
You really don’t want to miss Chi Feng‘s absolutely wonderful interactive demos. (1) Markov chain Monte Carlo sampling I believe this is exactly what Andrew was asking for a few Stan meetings ago: Chi Feng’s Interactive MCMC Sampling Visualizer This tool lets you explore a range of sampling algorithms including random-walk Metropolis, Hamiltonian Monte Carlo, […] The post Interactive visualizations…
I’ll be presenting at a webinar on Wednesday, December 13 at 1:00 PM Eastern. The title of the presentation is “Seven questions a statistician and answer for an attorney.” I will discuss, among other things, when common sense applies and when correct analysis can be counter-intuitive. There will be ample time at the end of […]
A commmenter points to a chapter of E. T. Jaynes’s book on probability and inference that contains the following amazing bit: The information we get from the TV evening news is not that a certain event actually happened in a certain way it is that some news reporter has claimed that it did. Even seeing […] The post Ed Jaynes…
Attention conservation notice: I have no taste. Oliver Morton, The Planet Remade: How Geoengineering Could Change the World This is just as impressive, enlightening, gracefully-written and thoughtful as I'd expect from the author of Eaters of the S...
There are important discussions nowadays about data modeling, to choose between the “two cultures” (as mentioned in Breiman (2001)), i.e. either econometrics models or machine/statistical learning models. We did discuss this issue recently in Econométrie et Machine Learning (so far only in French) with Emmanuel Flachaire and Antoine Ly. One argument often used by econometricians is the interpretability of econometric…
When using JAGS with censored data, the censored values are imputed to be consistent with the parameters of the model and the censoring limits. It's straight forward to record and graph the imputed values. Here are a couple of slides from my workshops ...
This week a neighbor of mine told me he wanted to learn R, and asked me how I learned R. I realized I had never blogged about it, so here we go. I learned R through a rather unusual way: I mostly learned R by answering other people’s questions. R was introduced to me in 2004 in a course on…
“Hey, remember me? I’ve been busy working like crazy” – Fever Ray I’m at the Banff International Research Station (BIRS) for the week, which is basically a Canadian version of Disneyland where during coffee breaks a Canadian woman with a rake politely walks around telling elk to “shoo”. The topic of this week’s workshop isn’t […] The post Always crashing…
Science writer Julie Rehmeyer discusses her own story: Five years ago, against practically anyone’s better judgment, I knowingly abandoned any semblance of medical evidence to follow the bizarre-sounding health advice of strangers on the internet. The treatment was extreme, expensive, and potentially dangerous. If that sounds like a terrible idea to you, imagine how it […] The post “How to…
I’m getting my computer repaired and so will be offline for a few days, won’t be reading or sending email or reading blog comments. The blog will auto-post, though, one per day, with already-scheduled material: “How to Assess Internet Cures Without Falling for Dangerous Pseudoscience” Ed Jaynes outta control! A reporter sent me a Jama […] The post Offline appeared…
Yu and Kumbier write: Artificial intelligence (AI) is intrinsically data-driven. It calls for the application of statistical concepts through human-machine collaboration during generation of data, development of algo- rithms, and evaluation of results. This paper discusses how such human-machine collaboration can be approached through the statistical concepts of population, question of interest, representativeness of training […] The post Bin Yu…
rOpenSci published an interview of Jenny Bryan today, and I enjoyed reading it. It is very much worth reading as a profile of an R developer, because it showed a few aspects and roles of a software developer that are often less known to the general public: a software developer as a (former) teacher, a researcher, a collaborator, and a…
This fascinating post by David Weakliem documents declining confidence in political institutions: and the news media: and some other institutions: As Weakliem writes: So far, confidence in everything has declined. You could offer specific explanations for each one, but the fact that it’s so widespread suggests that the declines reflect a general mood of dissatisfaction. […] The post Loss of…
Kaiser Fung, founder of Principal Analytics Prep and Columbia's MS in Applied Analytics, discusses an article by Rodney Brooks (MIT) on the future of AI.
Perhaps I have seen it before, but I just came across the “rubber duck debugging” method. It reminded me of a talk I gave at the 2013 BioC conference, in which I introduced the “marmot reproducible researching” method. That is, when you write code chunks in a knitr document, you need to explain to the marmot what you are doing…
Two years ago, I did a quick search on how to embed a live video through my webcam in my HTML5 presentations, and it turned out to be simple enough if you have a modern web browser. I was looking for this because sometimes when you do webinars, your audience may want to see you (perhaps to confirm that they…
“We become what we behold. We shape our tools and then our tools shape us.” — John Culkin. I came across this quote earlier this year, and I’d rate it as the most important single piece of thought I learned this year.1 This morning (actually last night) Karl Broman said on Twitter, “The tweets are too damn long.” And Christoffer…
“We become what we behold. We shape our tools and then our tools shape us.” — John Culkin. I came across this quote earlier this year, and I’d rate it as the most important single piece of thought I learned this year.1 This morning (actually last night) Karl Broman said on Twitter, “The tweets are too damn long.” And Christoffer…
The grand master of base R, Karl Broman, recently came up with an ingenious idea of showing missing values in scatterplots. Someone replied on Twitter that he had never seen it before. My PhD adviser Di replied never say never, and pointed out this idea had existed and been implemented for decades (e.g., in GGobi and MANET). It is a…
