Rachael Meager

@economeager

🏳️‍🌈🇦🇺 Assistant Prof . Development Econ & Econometrics. All views are purely my own as I am immune to propaganda. Bayesian menace and scallywag.

London via Boston & Melbourne
Vrijeme pridruživanja: svibanj 2010.

Tweetovi

Blokirali ste korisnika/cu @economeager

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @economeager

  1. Prikvačeni tweet
    10. lis 2017.

    Life motto from John Tukey

    Poništi
  2. proslijedio/la je Tweet

    In statistics and also in politics, if more people realised the 'the past is not on offer' analysis and decision-making would improve considerably.

    Poništi
  3. here's another paper recommended by that tackles the problem more directly but is much tougher to read, approach with enthusiasm for mathematics:

    Prikaži ovu nit
    Poništi
  4. It's been pointed out by that the wiki misleadingly makes it seem that the problem has been satisfyingly solved -- when it has not!! Check out this paper for more information and discussion:

    Prikaži ovu nit
    Poništi
  5. proslijedio/la je Tweet
    prije 46 minuta

    It is possible--again, stay with me here--that when candidates tweet predictions about election results, their goal isn't to minimize mean squared error.

    Poništi
  6. *does a little dance* doot doot doot if you look back over your notes very carefully you'll see your econometrics lecturers never promised you unbiased SE estimators, just consistent ones. doot doot doot.

    Prikaži ovu nit
    Poništi
  7. prije 1 sat

    "robust" (Eiker-Huber-White) style standard errors are substantially biased downwards in many cases - especially those with a few "high leverage" data points, which Young 2019 finds in much of applied work. See Chesher and Jewitt 1987 for the proof:

    Prikaži ovu nit
    Poništi
  8. prije 1 sat

    I've got biscuits in my office thank god

    Prikaži ovu nit
    Poništi
  9. prije 1 sat

    I think I hurt my hip while I was working out. I just stood up from the desk, felt the pain in my hip, and said the following things out loud to nobody in my office: "Owwww, far out, christ, fuck,....I need biscuits."

    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    prije 2 sata
    Odgovor korisniku/ci
    Poništi
  11. prije 2 sata

    the most interesting, confusing and beautiful things about statistics lie at its foundations and show themselves in its simplest operations. nothing can be taken for granted, one is plunged again and again into uncertainty. it's so exciting!

    Prikaži ovu nit
    Poništi
  12. prije 2 sata

    this has strong "legolas and gimli at the gate of the dead" energy

    Poništi
  13. proslijedio/la je Tweet
    prije 2 sata

    So is Stand Your Ground exclusively for shooting unarmed black people, or...?

    Poništi
  14. prije 4 sata

    Every now and then I forget and re-remember that the classical t-test of differences in means across groups assumes the variance of the two groups is the same, which is almost certainly problematic. The Behrens-Fisher problem! One of my faves:

    Prikaži ovu nit
    Poništi
  15. proslijedio/la je Tweet
    prije 21 sat

    If this does not become a plot point in a book or movie, I will be very disappointed.

    Poništi
  16. proslijedio/la je Tweet
    prije 23 sata

    The politeness rule--that if you can't say something positive, don't say anything at all--is close to reversed in intellectual contexts: if you don't have an objection, stay silent. So 'mixed' contexts (e.g. Twitter) create problems.

    Poništi
  17. proslijedio/la je Tweet

    See, people think Anscombe's quartet is all artificial extreme cases and no one would actually make such a stupid mistake with regression, and then you get figures like Fig 8C

    Prikaži ovu nit
    Poništi
  18. prije 24 sata

    as I have said before, 99% of everything is janky

    Poništi
  19. proslijedio/la je Tweet
    3. velj

    I’ve said this before on here but our criminal justice system is the literal plot of Les Miserables.

    Poništi
  20. proslijedio/la je Tweet
    2. velj

    One of the nice things about Andrew Gelman's blog is how benign posts routinely produce deep debates about foundations of statistics. This time, the concept of "calibration" (see comments): My attitude (new box in 2nd ed of my book):

    Prikaži ovu nit
    Poništi
  21. proslijedio/la je Tweet
    31. sij

    I remember some advice gave me to try to not be singularly focused on writing a job market paper, which encouraged me to revisit an old idea. That idea led to a better idea that eventually became my job market paper.

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·