Skip to primary content

Statistical Modeling, Causal Inference, and Social Science

Main menu

  • Home
  • Authors
  • Blogs We Read
  • Books
  • Sponsors

Post navigation

Getting into hot water over hot graphics A new efficient lossless compression algorithm

Visualization magazine

Posted on August 23, 2010 11:44 PM by Andrew

Aleks pointed me to this.

This entry was posted in Statistical Graphics by Andrew. Bookmark the permalink.
  • Art
  • Artificial Intelligence
  • Bayesian Statistics
  • Causal Inference
  • Decision Analysis
  • Economics
  • Jobs
  • Literature
  • Miscellaneous Science
  • Miscellaneous Statistics
  • Multilevel Modeling
  • Obituaries
  • Papers
  • Political Science
  • Public Health
  • Sociology
  • Sports
  • Stan
  • Statistical Computing
  • Statistical Graphics
  • Teaching
  • Zombies
  1. Seren Dipity on Blokjutsu: a macOS image generator/editorNovember 17, 2026 12:40 AM

    Bob, you and your “beer and pretzels” (need FFTESS ratings!*) childhood friend's will appreciate Simon Willison's prompt; "Then I tried telling…

  2. Ken Schulz on Thinking fast, slow, and not at all: System 3 jumps the sharkNovember 16, 2026 10:11 PM

    True; but torque on drive shafts is a better measure, especially if the road surface is wet, snowy or icy.…

  3. Andrew H. on Algorithmic deepfake scienceNovember 16, 2026 9:53 PM

    The biggest risk from algorithmic science is time spent verifying outputs. I think there are only two options: 1) automating…

  4. AAAnonymous on 7 steps to junk science that can achieve worldly successNovember 16, 2026 5:30 PM

    Quote from the blog post: "So here they are, 7 things that allow junk science to thrive:" “BRICKS that build…

  5. Meese on Algorithmic deepfake scienceNovember 16, 2026 4:56 PM

    Damn, that website is a giant plea for money. I also had never heard of it, I had to look…

  6. Tracy Lightcap on When is a class a “commodity” and when does it depend on the instructor?November 16, 2026 2:13 PM

    I looked at Active Statistics and liked it a lot. One complaint: a lack of coverage of tabular analysis. A…

  7. Tracy Lightcap on When is a class a “commodity” and when does it depend on the instructor?November 16, 2026 1:49 PM

    And this doesn't consider the plight of "methods" teachers. I was (sigh) a professor of political science at a small…

  8. Michael Murphy on Algorithmic deepfake scienceNovember 16, 2026 1:40 PM

    The Comet Research Group is a better example of this. The papers, from Younger Dryas Impact stuff on, are all…

  9. Anonymous on Algorithmic deepfake scienceNovember 16, 2026 1:04 PM

    Thanks for this discussion. I've read publications where LLMs were used to generate gynecologic medical statements, and it was (if…

  10. Anoneuoid on Algorithmic deepfake scienceNovember 16, 2026 12:44 PM

    If stages 1 and 2 are not met *independently of your research* in some way (doesnt matter how), I think…

  11. Anoneuoid on Algorithmic deepfake scienceNovember 16, 2026 12:31 PM

    Yes, of course NHST is a symptom. It could never have been so widely adopted without the transition to US…

  12. Dale Lehman on Algorithmic deepfake scienceNovember 16, 2026 12:14 PM

    I know you are fixated on NHST and I share many of your concerns with it. But, in this case,…

  13. Anoneuoid on Algorithmic deepfake scienceNovember 16, 2026 12:07 PM

    It is possible that this is a perceptual error on my part, but I feel like the ethical constraints on…

  14. Dale Lehman on Algorithmic deepfake scienceNovember 16, 2026 11:42 AM

    I don't really see the value of distinguishing between "fake" science and "deepfake" science. In both cases, the perpetrator is…

  15. Anonymous on Algorithmic deepfake scienceNovember 16, 2026 11:05 AM

    You can use the postmodernism generator (written 1996) to generate some plausible nonsense. https://www.elsewhere.org/journal/pomo/

  16. Anoneuoid on Algorithmic deepfake scienceNovember 16, 2026 11:02 AM

    Chatbots being used to create “deepfakes” of scientific research: papers that are professionally formatted, readable, and are written in the…

  17. Bob Carpenter on Algorithmic deepfake scienceNovember 16, 2026 11:01 AM

    In the U.S. going back as far as I got involved in the 1980s, academic hiring, tenure, and promotion is…

  18. Everson on OK, here’s a new scam. But I can’t figure out how it’s supposed to work.November 16, 2026 9:30 AM

    Hi Everyone, just sharing a similar version of the invite that I've received today. But they offered me only $1500…

  19. Ziggy on 2026 election conspiracies haven’t breached the Vermeule threshold.November 16, 2026 9:12 AM

    I'll try to defend the lawyers here, but not too much. Lawyers indeed have a tradition of advocacy, but they…

  20. Sean on 2026 election conspiracies haven’t breached the Vermeule threshold.November 15, 2026 10:50 PM

    Andrew, I see three common patterns: - the demagogue who can't resist explaining in front of a recorder that his…

  21. Andrew on Blokjutsu: a macOS image generator/editorNovember 15, 2026 6:00 PM

    Bob: In your post, you talk about "WotC slop," which makes me think about the general concept of "X slop,"…

  22. Andrew on Blokjutsu: a macOS image generator/editorNovember 15, 2026 5:47 PM

    Bob: You can post on non-statistical topics whenever you want! "Statistical Modeling, Causal Inference, and Social Science" is a broad…

  23. Joshua on 2026 election conspiracies haven’t breached the Vermeule threshold.November 15, 2026 4:56 PM

    John - Stop just criticizing the conclusion and not the method. On a stats blog no less. In 2026. https://statmodeling.stat.columbia.edu/2026/11/07/2026-compared-to-other-dramatic-midterm-election-reversals-of-the-past-half-century/#comment-2420488

  24. Andrew on 2026 election conspiracies haven’t breached the Vermeule threshold.November 15, 2026 3:52 PM

    John: Yes, indeed. As I've written elsewhere, the easiest way to lie with statistics is to just lie!

  25. John N-G on 2026 election conspiracies haven’t breached the Vermeule threshold.November 15, 2026 2:10 PM

    I know facts aren't exactly the topic of this blog post, but nonetheless: Number of counties in Texas where Trump…

  26. John N-G on 2026 election conspiracies haven’t breached the Vermeule threshold.November 15, 2026 1:55 PM

    p=1

  27. Andrew on 2026 compared to other dramatic midterm election reversals of the past half-centuryNovember 15, 2026 1:25 PM

    David: You write, "the swing in 1974 was a bigger swing than 2026." In terms of the vote, no, it…

  28. Dale Lehman on Zombie citations: A short empirical investigationNovember 15, 2026 12:58 PM

    For the most part I find the citations to be accurate - they were not a year ago but today…

  29. Andrew on 2026 election conspiracies haven’t breached the Vermeule threshold.November 15, 2026 12:50 PM

    Joshua: You might be right. I doubt I'll ever know. I've never met the man or corresponded with him, and,…

  30. Andrew on 2026 election conspiracies haven’t breached the Vermeule threshold.November 15, 2026 12:48 PM

    Oncodoc: Yes, in response to your question, "Is there a term that statisticians use for things that have actually happened":…

  31. oncodoc on 2026 election conspiracies haven’t breached the Vermeule threshold.November 15, 2026 12:38 PM

    Is there a term that statisticians use for things that have actually happened? Odds describe the likelihood of future events,…

  32. Joshua on 2026 election conspiracies haven’t breached the Vermeule threshold.November 15, 2026 12:27 PM

    Andrew - > He's well educated and, I assume, no idiot. You offer that as a reason Vermeule would know…

  33. Andrew on 2026 election conspiracies haven’t breached the Vermeule threshold.November 15, 2026 11:06 AM

    Gec: There is a logic to these spurious election-denial conspiracy theories. It goes like this: 1. The election did not…

  34. gec on 2026 election conspiracies haven’t breached the Vermeule threshold.November 15, 2026 10:04 AM

    Looking at the examples you shared, I was struck by how they demonstrate the logical issue Meehl and others identified…

  35. Ioannis Sagonas on When is a class a “commodity” and when does it depend on the instructor?November 15, 2026 6:38 AM

    Which book did you use to learn math stats back then and which one would you recommend today? I have…

  36. jcp on The challenge of truly probabilistic programmingNovember 14, 2026 10:20 PM

    This doesn't get you as far as you'd like, but I think working in Julia has some benefits in this…

  37. Joshua on If the Jensen Huang / Nvidia spy scandal had happened two weeks ago . . .November 14, 2026 10:09 PM

    Andrew - Perhaps not. And I certainly agree with your larger point: people tend to reverse engineer grand Just So…

  38. Joshua on If the Jensen Huang / Nvidia spy scandal had happened two weeks ago . . .November 14, 2026 10:02 PM

    Anoneuoid - You say this wasn't spontaneous, but you haven't said who did it, why, or what Jensen refused to…

  39. Raphael K on Zombie citations: A short empirical investigationNovember 14, 2026 5:28 PM

    @Raghu Pathasarathy: I am not sure when you last read a term paper written by a student at the lower…

  40. Kyle C on Zombie citations: A short empirical investigationNovember 14, 2026 5:21 PM

    That's interesting, in the law, judicial decisions about lawyers filing briefs containing hallucinated authorities are still on the upswing. I…

  41. Visruth on Zombie citations: A short empirical investigationNovember 14, 2026 5:09 PM

    Yeah supposedly not as big of a deal anymore: https://zou-group.github.io/llm-claims-shelf-life/ Very cool (interactive!) paper. See C11 for the citation result.

  42. Anonymous on If the Jensen Huang / Nvidia spy scandal had happened two weeks ago . . .November 14, 2026 4:33 PM

    Anoneuoid: A covert investigation of their own asset makes sense. An overt one that the media knows about does not.…

  43. Anoneuoid on If the Jensen Huang / Nvidia spy scandal had happened two weeks ago . . .November 14, 2026 3:04 PM

    @Anonymous The implausible premise was that the chinese gov was too incompetent to have open investigations into their intelligence assets.…

  44. Anonymous on If the Jensen Huang / Nvidia spy scandal had happened two weeks ago . . .November 14, 2026 2:43 PM

    It’s not a premise. It’s just a piece of supporting evidence. Maybe this wasn’t clear in the previous comment, I…

  45. Anoneuoid on If the Jensen Huang / Nvidia spy scandal had happened two weeks ago . . .November 14, 2026 2:12 PM

    Would Chinese intelligence need to investigate their own asset? Seems like the height of incompetence not to have ongoing investigations…

  46. Raghu Parthasarathy on Zombie citations: A short empirical investigationNovember 14, 2026 1:49 PM

    Are hallucinated citations still a major issue? In 2025, sure; that's when I think the data cited by the paper…

  47. Alper San on Know your data, recode missing data codesNovember 14, 2026 1:32 PM

    There is a second layer of missingness in time-series data: a missing field and a missing observation are different problems.…

  48. AAAnonymous on Zombie citations: A short empirical investigationNovember 14, 2026 11:57 AM

    This blog post reminded me of two articles I know of that mention some sort of checking of how citations…

  49. Anonymous on If the Jensen Huang / Nvidia spy scandal had happened two weeks ago . . .November 14, 2026 10:46 AM

    Right. Though in my view it's just that the neocons want to put more export controls on China. Jensen was…

  50. AAAnonymous on “I think there’s an argument to be made that much meta-scientific work is a kind of mirror image of the empirical work it critiques”November 14, 2026 10:36 AM

    Quote from above: "I think these two examples are relevant and possibly interesting examples concerning the quote above about meta-science.…

Proudly powered by WordPress