Skip to primary content

Statistical Modeling, Causal Inference, and Social Science

Main menu

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

Post navigation

R and Google Visualization Visual diagnostics for discrete-data regressions

Hey, look over here! Another rant!

Posted on September 27, 2011 9:11 AM by Andrew

Bigshot establishment dude Peter Orszag thinks bigshot establishment dudes don’t have enough power.

(Also politically related but not a rant:
Joe McCarthy Versus Powerman and the Debt-Ceiling Destroyers, Part One.)

This entry was posted in Political Science 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. Carlos Ungil on Posterior predictive checking is for non-Bayesians too!July 31, 2026 3:31 AM

    By the way, I suggested “parametric bootstrap” as a less wordy substitute for the “point estimate model simulation” part of…

  2. Steve Sailer on People sometimes talk about “the Jewish vote,” but what’s relevant is not really the Jewish vote or Jewish public opinion; it’s really about campaign contributions and the news media. Also similar with Mormons.July 30, 2026 10:10 PM

    October 7, 2023 appears to have had a big impact on rich Jewish donors. It's easy to look up the…

  3. Carlos Ungil on Posterior predictive checking is for non-Bayesians too!July 30, 2026 6:25 PM

    Ok, maybe we all agree (including our present and past selves) that when you sample using plug-in parameter estimates it…

  4. Andrew on Posterior predictive checking is for non-Bayesians too!July 30, 2026 6:10 PM

    Carlos: What I wrote then is consistent with what I am writing now. The parametric bootstrap is a technique that…

  5. Carlos Ungil on Posterior predictive checking is for non-Bayesians too!July 30, 2026 5:29 PM

    > My impression is that bootstrapping is usually considered as a method for getting the distribution of parameter estimates. In…

  6. confused on Out of the frying pan and into the fire: Scientific American returned to form, and then this happened:July 30, 2026 5:23 PM

    Yes, I think there's a very important difference between the well understood physics side of it (more greenhouse gases blocking…

  7. Anonymous AI slop policeman on “Over-coverage caught by pre-registration: 47 of 56 inside a stated 50% interval”July 30, 2026 5:21 PM

    The parts of this post attributed to Malinowski were in fact generated by AI. A nice test of Pangram’s new…

  8. confused on Out of the frying pan and into the fire: Scientific American returned to form, and then this happened:July 30, 2026 5:18 PM

    @Joshua: I fear that the necessary time lag between reduced CO2 emissions and temperature effects being visible means that "purely"…

  9. confused on Out of the frying pan and into the fire: Scientific American returned to form, and then this happened:July 30, 2026 5:13 PM

    I think more of them are honest (though wrong) than you think. The effects of global warming may be extremely…

  10. Carlos Ungil on Posterior predictive checking is for non-Bayesians too!July 30, 2026 4:46 PM

    > The “posterior” came in because . . . I was averaging over the posterior distribution of the parameter rather…

  11. Sung J. Woo on Eleven Kinds of Loneliness: Richard Yates and the tragedy of agencyJuly 30, 2026 4:43 PM

    Hello Andrew, always great to meet another Yates fan! Entirely too few of us, that's for sure. Drawing a parallel…

  12. Andrew on Posterior predictive checking is for non-Bayesians too!July 30, 2026 4:42 PM

    Carlos: My impression is that bootstrapping is usually considered as a method for getting the distribution of parameter estimates. In…

  13. Andrew on Posterior predictive checking is for non-Bayesians too!July 30, 2026 4:39 PM

    Carlos: As I wrote, it's a generalization of the plug-in approach. The "posterior" came in because . . . I…

  14. jd on Posterior predictive checking is for non-Bayesians too!July 30, 2026 3:50 PM

    Nice! Just found the same link and was about to post.

  15. Carlos Ungil on Posterior predictive checking is for non-Bayesians too!July 30, 2026 3:48 PM

    I think that "parametric bootstrap" is a usual term when you sample using plug-in parameter estimates. [ https://web.universiteitleiden.nl/fsw/verduin/stathist/1stword.htm ] POSTERIOR…

  16. jd on Posterior predictive checking is for non-Bayesians too!July 30, 2026 3:11 PM

    Yeah does seem like it would be ‘point estimate model simulation checking’ or something less wordy but similar… Does anyone…

  17. A frequent reader on Eleven Kinds of Loneliness: Richard Yates and the tragedy of agencyJuly 30, 2026 2:36 PM

    Thanks for clarifying! I can see how the bleakness can hit harder if the dark fate was, indeed, avoidable.

  18. Carlos Ungil on Posterior predictive checking is for non-Bayesians too!July 30, 2026 2:29 PM

    > at the time I was thinking of it more as a generalization of the existing “plug-in” approach that would…

  19. Andrew on Posterior predictive checking is for non-Bayesians too!July 30, 2026 2:20 PM

    Jd: From my perspective, the origins of posterior predictive checking are: 1. The chi-squared test of model fit. This came…

  20. Daniel Lakeland on He fit the same statistical models with three different software and got much different estimates. It’s another dimension of the multiverse.July 30, 2026 1:53 PM

    Dale, the article you mention seems like an attempt to provide the "guardrails" that were being promoted by our fellow…

  21. jd on Posterior predictive checking is for non-Bayesians too!July 30, 2026 1:50 PM

    Are the origins of posterior predictive checking, Frequentist residual analysis?

  22. Michael Minor on Posterior predictive checking is for non-Bayesians too!July 30, 2026 11:02 AM

    Industrial statistics is embracing the spirit of posterior predictive checks. For example, JMP 18 added visualization of prediction intervals and…

  23. Dale Lehman on “Archaeology can’t give social scientists population or GDP, but here are some things we can measure that might be useful for social science.”July 30, 2026 7:14 AM

    Somewhat along the lines of this post - and quite in line with many posts over the years - is…

  24. Seren Dipity on Eleven Kinds of Loneliness: Richard Yates and the tragedy of agencyJuly 30, 2026 1:20 AM

    Andrew, if you can cope - seems you can - and like a good movie vaugely in the vein of…

  25. Dale Lehman on Selection effects can go both ways (for taxi drivers as well as the rest of us)July 29, 2026 7:33 PM

    Thanks, you got it! Part of my memory is intact.

  26. Jonathan (another one) on Selection effects can go both ways (for taxi drivers as well as the rest of us)July 29, 2026 5:32 PM

    You're thinking of Sam Peltzman's work, pparticularly https://www.journals.uchicago.edu/doi/10.1086/260352 See also: https://www.econlib.org/what-the-peltzman-effect-is-and-isnt/

  27. Alex Malinowski on “Over-coverage caught by pre-registration: 47 of 56 inside a stated 50% interval”July 29, 2026 5:26 PM

    Author here. On the first point, which is the right one to press: yes, the conditioning variable is fixed before…

  28. shira on Survey Statistics: equivalent models, equivalent weights (locally)July 29, 2026 4:38 PM

    Thanks for sharing your experience, Raphael ! I started with very granular population data. I ended up using far more…

  29. Andrew on Eleven Kinds of Loneliness: Richard Yates and the tragedy of agencyJuly 29, 2026 4:37 PM

    Frequent: I see what you're saying, but I meant something slightly different. I guess I wasn't so clear. When I…

  30. A frequent reader on Eleven Kinds of Loneliness: Richard Yates and the tragedy of agencyJuly 29, 2026 4:23 PM

    "This is not a the sort of tragedy where the attempt to avert the prophecy leads to the prophecy being…

  31. Carlos Ungil on “Making Statistics Work: Information Theory and Bayesian Inference”July 29, 2026 2:45 PM

    > I don’t agree with much of it You gave as an example their discussion of "prior belief" which seems…

  32. Joshua on Out of the frying pan and into the fire: Scientific American returned to form, and then this happened:July 29, 2026 1:20 PM

    Anoneuoid - OK. This convo never had a huge change of being productive - but I will allow myself one…

  33. Chris on Out of the frying pan and into the fire: Scientific American returned to form, and then this happened:July 29, 2026 12:44 PM

    apols, didn't nest that properly - was a response to Anoneuoid

  34. Chris on Out of the frying pan and into the fire: Scientific American returned to form, and then this happened:July 29, 2026 12:40 PM

    In Europe Ursula von der Leyen, is a good example, also Teresa Ribera in Spain - perhaps women are less…

  35. Anoneuoid on Out of the frying pan and into the fire: Scientific American returned to form, and then this happened:July 29, 2026 11:11 AM

    @All Please tell me what any politician has done to stop warming and how we would determine the effectiveness of…

  36. Dale Lehman on He fit the same statistical models with three different software and got much different estimates. It’s another dimension of the multiverse.July 29, 2026 9:20 AM

    Daniel and others with similar concerns: The detail of this paper is not my skill set, so I am not…

  37. Joshua on Out of the frying pan and into the fire: Scientific American returned to form, and then this happened:July 29, 2026 9:14 AM

    Dale - Of course he was being sarcastic about the Trump "fighting" global warming part. I certainly hope he was…

  38. Dale Lehman on Out of the frying pan and into the fire: Scientific American returned to form, and then this happened:July 29, 2026 9:03 AM

    I thought Anoneuoid was being purposefully sarcastic (though you can never know). I think the point about the actual effects…

  39. Joshua on Out of the frying pan and into the fire: Scientific American returned to form, and then this happened:July 29, 2026 8:46 AM

    Anoneuoid – > "Neither seemed to have any effect." It would be absurd to expect to see a measurable "effect"…

  40. T.Bickle on Selection effects can go both ways (for taxi drivers as well as the rest of us)July 29, 2026 8:06 AM

    I drove a cab to get through college. You want the driver who wears their seatbelt because much of what…

  41. David, a Bostonian in Tokyo on How generic language shapes the development of social thoughtJuly 29, 2026 8:04 AM

    My reading of Whorf-Sapir was they wanted to argue people in different cultures were substantially/significantly different in their thought processes,…

  42. Michael L. on Selection effects can go both ways (for taxi drivers as well as the rest of us)July 29, 2026 8:00 AM

    Like Dale (if I may), I don’t see the connection in the seat belt example to adverse selection. So I…

  43. Raphael K on Survey Statistics: equivalent models, equivalent weights (locally)July 29, 2026 7:23 AM

    By all means I am no expert in adjusting for the population distribution. I tried it for the first time…

  44. Jeremy Miles on Selection effects can go both ways (for taxi drivers as well as the rest of us)July 28, 2026 11:33 PM

    At least in the UK, there's another dimensions - safety issues. Taxi drivers there don't need to use a seatbelt…

  45. Savio Mak on Selection effects can go both ways (for taxi drivers as well as the rest of us)July 28, 2026 11:32 PM

    so, one man's modus ponens is another man's modus tollens

  46. Anoneuoid on Out of the frying pan and into the fire: Scientific American returned to form, and then this happened:July 28, 2026 11:04 PM

    Between covid and iran, Trump has done more to fight global warming than any other politician. First demand destruction, then…

  47. Dale Lehman on Selection effects can go both ways (for taxi drivers as well as the rest of us)July 28, 2026 7:18 PM

    Rob Thanks for the interesting references. I don't think that was the source I had in mind, but it suggests…

  48. Rob Searle on Selection effects can go both ways (for taxi drivers as well as the rest of us)July 28, 2026 5:45 PM

    Dale - I think that you are looking for the (many) publications of Professor John Adams and his concept of…

  49. shira on Survey Statistics: equivalent models, equivalent weights (locally)July 28, 2026 4:55 PM

    You're right, Andrew ! I meant to honor Basu by extending his example to Bears, so I rephrased.

  50. Andrew on Survey Statistics: equivalent models, equivalent weights (locally)July 28, 2026 4:41 PM

    Shira: I think that, out of respect for Basu, who came up with the example, you should keep his animal.

Proudly powered by WordPress