Jonathan Roth

@jondr44

I study econometrics econ. RT ≠ endorsement. Insignificant pretrend ≠ valid research design.

Vrijeme pridruživanja: svibanj 2011.

Tweetovi

Blokirali ste korisnika/cu @jondr44

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @jondr44

  1. proslijedio/la je Tweet

    It's a sign of how crazy the news cycle is right now that the United States three days ago barred immigration from the largest economy in Africa--a quarter of the continent's population is now barred from the US--and almost nobody is talking about it.

    Poništi
  2. proslijedio/la je Tweet

    I summarize three approaches: i) and on robustness to violations of a given magnitude; ii) & on how your base specification should allow for non-parallel trends; & et al.'s AER paper on using an IV-like approach

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet

    Today in part 2 of my attempt to catch up on the new DiD literature: Revisiting the Difference-in-Differences Parallel Trends Assumption: Part II What happens if the parallel trends assumption is (might be) violated?

    Prikaži ovu nit
    Poništi
  4. 2. velj

    On today's episode of "Correlation or Causation"...

    Poništi
  5. proslijedio/la je Tweet
    31. sij
    Odgovor korisnicima i sljedećem broju korisnika:

    Aren't you at a flyout dinner? 😂 If not, where is my husband????

    Poništi
  6. proslijedio/la je Tweet
    29. sij

    Anyone having trouble sorting through the hundreds (possibly millions?) of papers on advanced DiD methods floating around today? Come to Berkeley next summer and make me do the work for you.

    Prikaži ovu nit
    Poništi
  7. 22. sij

    Thanks to for this nice blog post discussing mine and others' work on pre-trends testing!

    Poništi
  8. proslijedio/la je Tweet

    Is your difference-in-differences reading list also getting really long? In today's post I summarize several papers that revisit the parallel trends assumption and what we learn from testing it holds in the pre-intervention periods

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

    Since it is Monday, it's time for the 'metrics paper of the week! Today I want to highlight "Bounds on Distributional Treatment Effect Parameters using Panel Data with an Application on Job Displacement", by Brantly Callaway. Link here:

    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    13. sij

    For the next month or so, I plan to mention an Econometrics paper that I really enjoy reading. I will try to do this every Monday. The first paper of this series is by and , "An Honest Approach to Parallel Trends" Link here:

    Prikaži ovu nit
    Poništi
  11. 2. sij

    Hey - my new strategy for not being an et al is to stand around these booths and try to make friends 😅

    Poništi
  12. proslijedio/la je Tweet

    dplyr::across() is now available in the dev version of . across() can be used inside mutate() or summarise() when you want to apply one (or several) function(s) to a selection of columns.

    Prikaži ovu nit
    Poništi
  13. 23. pro 2019.
    Poništi
  14. 23. pro 2019.

    The real Q is how do you play dreidel in space

    Poništi
  15. proslijedio/la je Tweet
    21. pro 2019.

    This can be a hard time of year for many folks. The National Suicide Prevention Lifeline is 1-800-273-8255. A simple copy and paste might save someone’s life. Would 3 Twitter friends please copy this text and post under their own name?

    Poništi
  16. 18. pro 2019.

    I take this as the highest of compliments 😃😃😃

    Poništi
  17. proslijedio/la je Tweet
    12. pro 2019.

    A thought on talks that I've heard repeatedly and really like. Suppose you're presenting something involved—e.g., where results rely on a few layers of definitions or intricate notation. (Almost everything at the research frontier is involved in some way.) 1/7

    Prikaži ovu nit
    Poništi
  18. proslijedio/la je Tweet

    how come this isnt impeachable, i wanna know

    Poništi
  19. proslijedio/la je Tweet
    3. pro 2019.

    Fellow JMCs: here is the walking travel time between ASSA hotels. Let me know if you notice any errors. Coordinates found with googleplaces Stata package and travel times computed with osrmtime.

    Prikaži ovu nit
    Poništi
  20. proslijedio/la je Tweet
    28. stu 2019.

    “An honest approach to parallel trends” by Rambachan and Roth (Roth’s JMP).

    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:

    ·