Published or to be published articles:

- [2021] Reflections on Lakatos’s “Proofs and Refutations.” {\em American Mathematical Monthly}. (Andrew Gelman)
- [2021] Holes in Bayesian statistics. {\em Journal of Physics G: Nuclear and Particle Physics}. (Andrew Gelman and Yuling Yao)
- [2021] Reflections on Breiman’s Two Cultures of Statistical Modeling. {\em Observational Studies}. (Andrew Gelman)
- [2021] Bayesian statistics and modelling. {\em Nature Reviews}. (Rens van de Schoot, Sarah Depaoli, Ruth King, Bianca Kramer, Kaspar Märtens, Mahlet G. Tadesse, Marina Vannucci, Andrew Gelman, Duco Veen, Joukje Willemsen, and Christopher Yau)
- [2021] Community prevalence of SARS-CoV-2 in England: Results from the ONS Coronavirus Infection Survey Pilot. {\em Lancet Public Health}. (Koen B. Pouwels, Thomas House, Emma Pritchard, Julie V. Robotham, Paul J. Birrell, Andrew Gelman, Karina-Doris Vihta, Nikola Bowers, Ian Boreham, Heledd Thomas, James Lewis, Iain Bell, John I. Bell, John N. Newton, Jeremy Farrar, Ian Diamond, Pete Benton, Ann Sarah Walker, and the COVID-19 Infection Survey Team)
- [2020] Information, incentives, and goals in election forecasts. {\em Judgment and Decision Making} {\bf 15}, 863–880. (Andrew Gelman, Jessica Hullman, Christopher Wlezien, and George Elliott Morris)
- [2020] An updated dynamic Bayesian forecasting model for the 2020 election. {\em Harvard Data Science Review}. (Merlin Heidemanns, Andrew Gelman, and Elliott Morris)
- [2020] Know your population and know your model: Using model-based regression and poststratification to generalize findings beyond the observed sample. {\em Psychological Methods}. (Lauren Kennedy and Andrew Gelman)
- [2020] Bayesian hierarchical weighting adjustment and survey inference. {\em Survey Methodology}. (Yajuan Si, Rob Trangucci, Jonah Gabry, and Andrew Gelman)
- [2020] Bayesian analysis of tests with unknown specificity and sensitivity. {\em Journal of the Royal Statistical Society C, Applied Statistics} {\bf 69}, 1269–1284. (Andrew Gelman and Bob Carpenter)
- [2020] Fallout of lead over Paris from the 2019 Notre-Dame cathedral fire. {\em GeoHealth}. (Alexander van Geen, Yuling Yao, Tyler Ellis, and Andrew Gelman)
- [2020] Evidence vs.\ truth. {\em Chance}. (Andrew Gelman)
- [2020] Improving multilevel regression and poststratification with structured priors. {\em Bayesian Analysis}.

(Yuxiang Gao, Lauren Kennedy, Daniel Simpson, and Andrew Gelman) - [2020] Using Bayesian analysis to account for uncertainty and adjust for bias in coronavirus sampling. {\em International Society for Bayesian Analysis Bulletin} {\bf 27} (2), 11–12. (Andrew Gelman and Bob Carpenter)
- [2020] Rank-normalization, folding, and localization: An improved R-hat for assessing convergence of MCMC. {\em Bayesian Analysis}. (Aki Vehtari, Andrew Gelman, Daniel Simpson, Bob Carpenter, and Paul-Christian Bürkner)
- [2020] Data visualization as narrative. {\em Frieze}. (Andrew Gelman and Helen DeWitt)
- [2020] Lessons learned and remaining challenges for online seminars and conferences. {\em Amstat News}. (Lauren Kennedy, Guillaume Basse, Andrew Gelman, Guido Imbens, Yajuan Si, Dominik Rothenhausler, and Jann Spiess)
- [2020] Expectation propagation as a way of life: A framework for Bayesian inference on partitioned data. {\em Journal of Machine Learning Research} {\bf 21}, 1–53. (Aki Vehtari, Andrew Gelman, Tuomas Sivula, Pasi Jylanki, Dustin Tran, Swupnil Sahai, Paul Blomstedt, John P. Cunningham, David Schiminovich, and Christian P. Robert)
- [2020] Laplace’s theories of cognitive illusions, heuristics, and biases (with discussion). {\em Statistical Science} {\bf 35}, 159–170. (Joshua B. Miller and Andrew Gelman)

Rejoinder to discussion of Laplace’s theories of cognitive illusions, heuristics, and biases. {\em Statistical Science} {\bf 35}, 175–177. (Joshua B. Miller and Andrew Gelman) - [2020] Statistics as squid ink: How prominent researchers can get away with misrepresenting data. {\em Chance}. (Andrew Gelman and Alexey Guzey)
- [2020] Voter registration databases and MRP: Toward the use of large scale databases in public opinion research. {\em Political Analysis}. (Yair Ghitza and Andrew Gelman)
- [2020] A consensus-based transparency checklist. {\em Nature Human Behaviour} {\bf 4}, 561–563. (Balazs Aczel, Barnabas Szaszi, Alexandra Sarafoglou, Zoltan Kekecs, Šimon Kucharský, Daniel Benjamin, Christopher Chambers, Agneta Fisher, Andrew Gelman, et al.)

Unpublished articles:

- Inference from non-random samples using Bayesian machine learning (Yutao Liu, Andrew Gelman, and Qixuan Chen)
- What are the most important statistical ideas of the past 50 years? (Andrew Gelman and Aki Vehtari)
- A proposal for informative default priors scaled by the standard error of estimates (Erik van Zwet and Andrew Gelman)
- A fast linear regression via SVD and marginalization. (Philip Greengard, Andrew Gelman, and Aki Vehtari)
- The piranha problem: Large effects swimming in a small pond. (Christopher Tosh, Philip Greengard, Ben Goodrich, Andrew Gelman, and Daniel Hsu)
- Bayesian workflow. (Andrew Gelman, Aki Vehtari, Daniel Simpson, Charles C. Margossian, Bob Carpenter, Yuling Yao, Paul-Christian Bürkner, Lauren Kennedy, Jonah Gabry, Martin Modrák)
- Interactive analysis needs theories of inference. (Jessica Hullman and Andrew Gelman)
- Using sex and gender in survey adjustment. (Lauren Kennedy, Katharine Khanna, Daniel Simpson, and Andrew Gelman)
- Adaptive path sampling in metastable posterior distributions. (Yuling Yao, Collin Cademartori, Aki Vehtari, and Andrew Gelman)
- A generational voting model for forecasting the 2020 American presidential election. (Jonathan Auerbach, Yair Ghitza, and Andrew Gelman)
- Build your own statistics course for students in a non-quantitative field. (Alexis Lerner and Andrew Gelman)
- Stacking for non-mixing Bayesian computations: The curse and blessing of multimodal posteriors. (Yuling Yao, Aki Vehtari, and Andrew Gelman)
- Accounting for uncertainty during a pandemic. (Jon Zelner, Julien Riou, Ruth Etzioni, and Andrew Gelman)
- A simple explanation for declining temperature sensitivity with warming. (E. M. Wolkovich, J. L. Auerbach, C. J. Chamberlain, D. M. Buonaiuto, A. K. Ettinger, I. Morales-Castilla, and A. Gelman)
- Reconciling evaluations of the Millennium Villages Project. (Andrew Gelman, Shira Mitchell, Jeffrey D. Sachs, and Sonia Sachs)
- How to embrace variation and accept uncertainty in linguistic and psycholinguistic data analysis.

(Shravan Vasishth and Andrew Gelman) - Validating Bayesian inference algorithms with simulation-based calibration. (Sean Talts, Michael Betancourt, Daniel Simpson, Aki Vehtari, and Andrew Gelman)
- Slamming the sham: A Bayesian model for adaptive adjustment with noisy control data. (Andrew Gelman and Matthijs Vákár)

Book:

Thank you so much, Aki, Aileen, Alex, Alexey, Alexis, Balazs, Ben, Bob, Charles, Chris, Chris, Christian, Collin, Dan, Daniel, David, Dominik, Dustin, Elliott, Erik, Guido, Guillaume, Helen, Jann, Jessica, Jon, Jonah, Jonathan, Jeff, Jennifer, John, Josh, Julien, Kate, Koen, Lauren, Lex, Lizzie, Martin, Matthijs, Merlin, Michael, Pasi, Paul, Paul, Philip, Qixuan, Rob, Ruth, Sean, Shira, Shravan, Sonia, Swupnil, Tuomas, Yair, Yajuan, Yuling, Yutao, and journal editors and reviewers—and lots more collaborators including Ben, Bryn, Erik, Gustavo, Hagai, Len, Lu, Margaret, Manu, Michael, Mitzi, Rachael, Shiro, Siquan, Shiro, Steve, Steve, Susan, Tamara, Tom, Vivienne, Witold, Yotam, and others on ongoing projects that we haven’t yet written up or published. And thanks to all our collaborators of the past decade and to our blog authors and commenters—especially the ones who disagreed with us. And to all you lurkers out there who read the blog faithfully but haven’t found the need to comment yet. And even to the annoying people out there who misrepresented us or presented flat-out bad analyses: you kept us on our toes! And to the developers of the internet and maintainers of the WordPress software and the IT team at Columbia University. We also thank our closest collaborator, and all the 200-year-old mentors out there. And our predecessors and contemporaries who changed statistics in so many ways during the past half century. And thanks to Laura Dickson’s 8th grade students at Sky Ranch Middle School in Sparks, Nevada, who interviewed me for their class project. And to everyone who wrote everything we read this year. And to the developers of vaccines, the poll workers, the growers and developers of food, and everyone else who kept the world going for all of us. And to our loved ones, and to the people who intersected our lives in less pleasant ways as well. And to Mary Gray who was so supportive when I took that class many years ago, and to Grace Yang who taught me stochastic processes a few years after that. I couldn’t follow half of what was going on—I’d just keep scribbling in my notebook beyond whatever I could understand—but she conveyed a sense of the mathematical excitement of the topic.

Very Impressive Andrew. Look forward to more discussions. A Productive and Safe New Year to you and all the interesting bloggers here.

What about the cats?

+1

Happy New Year Andrew. Thank you for everything you do!

Can I say for all of us who studied a few static subjects at Uni many thanks for your blog. It is highly informative and at times quite funny

Are you going to continue to speak out against White Supremacy and in support of Racial Justice in 2021, or has those problem been solved with the cancellation of Mr. Trump?

Somehow I missed the Chance article on Truth vs Evidence (I ended my membership to ASA, sticking only to RSS, perhaps a mistake in retrospect, as I only get Significance now).

Here’s an interesting thing you write in that article:

“It should be possible to say that your data are consistent with your prior beliefs while acknowledging that these data also admit other interpretations”

I was talking to Michael Betancourt about this point the other day as well. I’ve tried this in four recent papers, saying that our statistical modeling shows that the data are consistent with a certain theoretical position, but could have other interpretations. The end-result was either a desk reject or a rejection, on grounds that no decisive conclusion was possible from the data. The implication is clear: only papers that present decisive conclusions constitute an advance in knowledge and are therefore admissible as articles.

Michael has an interesting suggestion: write two versions of the article. One, on arXiv or wherever, tells it like it really is. The other is for editor/reviewer consumption, making decisive claims. Quite an interesting proposal, which I plan to try out. One could then put up a notice next to the published paper: “For a more statistically defensible version of this paper, please read the arXiv edition.”

Your university PR department would be unhappy with you.

I think I am safe from the PR dept of my uni.

Shravan said,

“The end-result was either a desk reject or a rejection, on grounds that no decisive conclusion was possible from the data. The implication is clear: only papers that present decisive conclusions constitute an advance in knowledge and are therefore admissible as articles.”

Aargh! People who can’t accept uncertainty as just a normal part of reality are — well, off in a fairy tale world.

+1

Nice reminiscence on Proofs and Refutations. It had a very similar effect on me, who knew early on that I was never going to be a mathematician, settling for the mathematics-adjacent economics, and constantly battling, mostly unsuccessfully, over a long career to impress on people just how special (as opposed to general) the big theorems of economics were. [Deep economics secret: whenever you meet an honest economist who has given you some deep insight into the workings of markets or institutions or people and tells you there are theorems to back the insights up, watch their face turn red when asked “Is that still true in the presence of sufficiently high transactions costs?”)

It is one thing to lack generality when discussing icosahedra; it is another thing altogether when discussing the choice behavior of human beings. All of the failings of game theory in economics, IMO, stem directly from the theorems that are provable knocking fruitlessly at the actual door leading to human interactions.

2020 was my first year reading and commenting on this blog; thanks to everyone who contributed and commented, and especially to Andrew who organizes the whole thing! It’s been quite interesting and thought-provoking.

I found it difficult to track specific conversations in the comments; how do you guys do it?

(If there was a way to sort the comments on a blog post by time posted, that would help.)

“I found it difficult to track specific conversations in the comments; how do you guys do it?

(If there was a way to sort the comments on a blog post by time posted, that would help.)”

The right side bar goes from least recent at the bottom to most recent at the top, so that helps (although not perfect).

Yes,I know, but I check so infrequently that most comments that are new to me are not listed on it.