Statistical computing – StatsBlogs

All I need is time, a moment that is mine, while I’m in between

Dan Simpson — Sat, 29 Jun 2019 02:14:41 +0000

You’re an ordinary boy and that’s the way I like it – Magic Dirt Look. I’ll say something now, so it’s off my chest. I hate order statisics. I loathe them. I detest them. I wish them nothing but ill and strife. They are just awful. And I’ve spent the last god only knows how long […]

The post All I need is time, a moment that is mine, while I’m in between appeared first on StatsBlogs.

How does Stan work? A reading list.

Andrew — Wed, 26 Jun 2019 13:14:46 +0000

Bob writes, to someone who is doing work on the Stan language: The basic execution structure of Stan is in the JSS paper (by Bob Carpenter, Andrew Matt Hoffman, Daniel Lee, Ben Goodrich, Michael Betancourt, Marcus Brubaker, Jiqiang Guo, Peter Li, and Allen Riddell) and in the reference manual. The details of autodiff are in […]

The post How does Stan work? A reading list. appeared first on StatsBlogs.

AnnoNLP conference on data coding for natural language processing

Bob Carpenter — Tue, 25 Jun 2019 19:00:16 +0000

This workshop should be really interesting: Aggregating and analysing crowdsourced annotations for NLP EMNLP Workshop. November 3–4, 2019. Hong Kong. Silviu Paun and Dirk Hovy are co-organizing it. They’re very organized and know this area as well as anyone. I’m on the program committee, but won’t be able to attend. I really like the problem […]

The post AnnoNLP conference on data coding for natural language processing appeared first on StatsBlogs.

How to simulate an instrumental variables problem?

Andrew — Fri, 21 Jun 2019 02:32:49 +0000

Edward Hearn writes: In an effort to buttress my own understanding of multi-level methods, especially pertaining to those involving instrumental variables, I have been working the examples and the exercises in Jennifer Hill’s and your book. I can find general answers at the Github repo for ARM examples, but for Chapter 10, Exercise 3 (simulating […]

The post How to simulate an instrumental variables problem? appeared first on StatsBlogs.

New! from Bales/Pourzanjani/Vehtari/Petzold: Selecting the Metric in Hamiltonian Monte Carlo

Andrew — Fri, 31 May 2019 23:29:25 +0000

Ben Bales, Arya Pourzanjani, Aki Vehtari, Linda Petzold write: We present a selection criterion for the Euclidean metric adapted during warmup in a Hamiltonian Monte Carlo sampler that makes it possible for a sampler to automatically pick the metric based on the model and the availability of warmup draws. Additionally, we present a new adaptation […]

The post New! from Bales/Pourzanjani/Vehtari/Petzold: Selecting the Metric in Hamiltonian Monte Carlo appeared first on StatsBlogs.

Neural nets vs. regression models

Andrew — Tue, 21 May 2019 13:55:46 +0000

Eliot Johnson writes: I have a question concerning papers comparing two broad domains of modeling: neural nets and statistical models. Both terms are catch-alls, within each of which there are, quite obviously, multiple subdomains. For instance, NNs could include ML, DL, AI, and so on. While statistical models should include panel data, time series, hierarchical […]

The post Neural nets vs. regression models appeared first on StatsBlogs.

Maintenance cost is quadratic in the number of features

Andrew — Fri, 10 May 2019 13:56:39 +0000

Bob Carpenter shares this story illustrating the challenges of software maintenance. Here’s Bob: This started with the maintenance of upgrading to the new Boost version 1.69, which is this pull request: https://github.com/stan-dev/math/pull/1082 for this issue: https://github.com/stan-dev/math/issues/1081 The issue happens first, then the pull request, then the fun of debugging starts. Today’s story starts an issue […]

The post Maintenance cost is quadratic in the number of features appeared first on StatsBlogs.

Stan examples in Harezlak, Ruppert and Wand (2018) Semiparametric Regression with R

Bob Carpenter — Mon, 06 May 2019 18:00:17 +0000

I saw earlier drafts of this when it was in preparation and they were great. Jarek Harezlak, David Ruppert and Matt P. Wand. 2018. Semiparametric Regression with R. UseR! Series. Springer. I particularly like the careful evaluation of variational approaches. I also very much like that it’s packed with visualizations and largely based on worked […]

The post Stan examples in Harezlak, Ruppert and Wand (2018) Semiparametric Regression with R appeared first on StatsBlogs.

Several post-doc positions in probabilistic programming etc. in Finland

Aki Vehtari — Thu, 25 Apr 2019 15:38:38 +0000

There are several open post-doc positions in Aalto and University of Helsinki in 1. probabilistic programming, 2. simulator-based inference, 3. data-efficient deep learning, 4. privacy preserving and secure methods, 5. interactive AI. All these research programs are connected and collaborating. I (Aki) am the coordinator for the project 1 and contributor in the others. Overall […]

The post Several post-doc positions in probabilistic programming etc. in Finland appeared first on StatsBlogs.

“Sometimes all we have left are pictures and fear”: Dan Simpson talk in Columbia stat dept, 4pm Monday

Andrew — Thu, 18 Apr 2019 00:16:29 +0000

4:10pm Monday, April 22 in Social Work Bldg room 903: Data is getting weirder. Statistical models and techniques are more complex than they have ever been. No one understand what code does. But at the same time, statistical tools are being used by a wider range of people than at any time in the past. […]

The post “Sometimes all we have left are pictures and fear”: Dan Simpson talk in Columbia stat dept, 4pm Monday appeared first on StatsBlogs.

What is the most important real-world data processing tip you’d like to share with others?

Andrew — Wed, 10 Apr 2019 15:30:41 +0000

This question was in today’s jitts for our communication class. Here are some responses: Invest the time to learn data manipulation tools well (e.g. tidyverse). Increased familiarity with these tools often leads to greater time savings and less frustration in future. Hmm it’s never one tip.. I never ever found it useful to begin writing […]

The post What is the most important real-world data processing tip you’d like to share with others? appeared first on StatsBlogs.

Markov chain Monte Carlo doesn’t “explore the posterior”

Bob Carpenter — Mon, 25 Mar 2019 18:52:30 +0000

First some background, then the bad news, and finally the good news. Spoiler alert: The bad news is that exploring the posterior is intractable; the good news is that we don’t need to. Sampling to characterize the posterior There’s a misconception among Markov chain Monte Carlo (MCMC) practitioners that the purpose of sampling is to […]

The post Markov chain Monte Carlo doesn’t “explore the posterior” appeared first on StatsBlogs.

Yes, I really really really like fake-data simulation, and I can’t stop talking about it.

Andrew — Sat, 23 Mar 2019 13:26:02 +0000

Rajesh Venkatachalapathy writes: Recently, I had a conversation with a colleague of mine about the virtues of synthetic data and their role in data analysis. I think I’ve heard a sermon/talk or two where you mention this and also in your blog entries. But having convinced my colleague of this point, I am struggling to […]

The post Yes, I really really really like fake-data simulation, and I can’t stop talking about it. appeared first on StatsBlogs.

My two talks in Montreal this Friday, 22 Mar

Andrew — Wed, 20 Mar 2019 15:00:40 +0000

McGill University Biostatistics seminar, Purvis Hall, 102 Pine Ave. West, Room 25, 1-2pm Fri 22 Mar: Resolving the Replication Crisis Using Multilevel Modeling In recent years we have come to learn that many prominent studies in social science and medicine, conducted at leading research institutions, published in top journals, and publicized in respected news outlets, […]

The post My two talks in Montreal this Friday, 22 Mar appeared first on StatsBlogs.

stanc3: rewriting the Stan compiler

Sean Talts — Wed, 13 Mar 2019 19:49:23 +0000

I’d like to introduce the stanc3 project, a complete rewrite of the Stan 2 compiler in OCaml. Join us! With this rewrite and migration to OCaml, there’s a great opportunity to join us on the ground floor of a new era. Your enthusiasm for or expertise in programming language theory and compiler development can help […]

The post stanc3: rewriting the Stan compiler appeared first on StatsBlogs.

HMC step size: How does it scale with dimension?

Andrew — Sat, 23 Feb 2019 14:17:33 +0000

A bunch of us were arguing about how the Hamiltonian Monte Carlo step size should scale with dimension, and so Bob did the Bob thing and just ran an experiment on the computer to figure it out. Bob writes: This is for standard normal independent in all dimensions. Note the log scale on the x […]

The post HMC step size: How does it scale with dimension? appeared first on StatsBlogs.

R fixed its default histogram bin width!

Andrew — Mon, 18 Feb 2019 14:54:46 +0000

I remember hist() in R as having horrible defaults, with the histogram bars way too wide. (See this discussion: A key benefit of a histogram is that, as a plot of raw data, it contains the seeds of its own error assessment. Or, to put it another way, the jaggedness of a slightly undersmoothed histogram […]

The post R fixed its default histogram bin width! appeared first on StatsBlogs.

Book reading at Ann Arbor Meetup on Monday night: Probability and Statistics: a simulation-based introduction

Bob Carpenter — Thu, 14 Feb 2019 20:00:46 +0000

The Talk I’m going to be previewing the book I’m in the process of writing at the Ann Arbor R meetup on Monday. Here are the details, including the working title: Probability and Statistics: a simulation-based introduction Bob Carpenter Monday, February 18, 2019 Ann Arbor SPARK, 330 East Liberty St, Ann Arbor I’ve been to […]

The post Book reading at Ann Arbor Meetup on Monday night: Probability and Statistics: a simulation-based introduction appeared first on StatsBlogs.

Should he go to grad school in statistics or computer science?

Andrew — Tue, 12 Feb 2019 14:36:31 +0000

Someone named Nathan writes: I am an undergraduate student in statistics and a reader of your blog. One thing that you’ve been on about over the past year is the difficulty of executing hypothesis testing correctly, and an apparent desire to see researchers move away from that paradigm. One thing I see you mention several […]

The post Should he go to grad school in statistics or computer science? appeared first on StatsBlogs.

Autodiff! (for the C++ jockeys in the audience)

Andrew — Wed, 30 Jan 2019 14:33:41 +0000

Here’s a cool thread from Bob Carpenter on the Stan forums, “A new continuation-based autodiff by refactoring,” on the inner workings of Stan. Enjoy.
P.S. Just for laffs, see this post from 2010 which is the earliest mention of auto...

The post Autodiff! (for the C++ jockeys in the audience) appeared first on StatsBlogs.