PostDoc, Research Network Affiliate at CESifo Munich
Stockholm, Sweden
Joined March 2012
- Tweets50
- Following173
- Followers107
- Likes131
Sven Resnjanskij retweeted
A German mentoring programme markedly boosts school achievement & labour market outcomes of students from disadvantaged backgrounds.
@SvenRes @unierfurt, @JRuhose @kieluni, @katharina_wedel @LMU_Muenchen @SimonWiederhold @IWH_Halle @Woessmann @ifo_Institut
ow.ly/X2ms50QjNes
Sven Resnjanskij retweeted
Mentoring improves the school-to-work transition of disadvantaged adolescents
@SvenRes @unierfurt, @JRuhose @kieluni, @katharina_wedel @LMU_Muenchen, @SimonWiederhold @UniHalle, @Woessmann @ifo_Institut
ow.ly/vg1m50QjNOr
Sven Resnjanskij retweeted
🚀Mentoring durch ehrenamtliche Studierende verdoppelt ‼️ die Chance von benachteiligten Jugendlichen, eine Berufsausbildung zu beginnen👍
Unsere Evaluierung der Effekte von @RYL_Mentoring
Ergebnisse jetzt im @ifo_Institut Schnelldienst:
👉ifo.de/publikationen/2023/ze…
🧵 1/6
Sven Resnjanskij retweeted
🚨Now "just accepted" @JPolEcon:
Can Mentoring Alleviate Family Disadvantage in Adolescence? A Field Experiment to Improve Labor-Market Prospects
w/ @SvenRes @JRuhose @SimonWiederhold @katharina_wedel
👉Strong 3-year effects of mentoring RCT
journals.uchicago.edu/doi/10…
🧵 1/9
The bike ride was great fun! Thanks, Josh! Next ride in Stockholm, then? @metrics52
Sven Resnjanskij retweeted
😎Thanks everyone for a terrific conference on school reforms.🎉This was truly outstanding!🥳
@metrics52 @loeb_susanna @edwinleuven @MariSRege @hooster1 @pqblair @zbleemer @LarissaZierow @Gabri_EllaConti @Bjorn_Ockert @erikgronqvist @m_silliman @urs_schaede @McmSandi …
Sven Resnjanskij retweeted
After 5 yrs of iterating on math, simulations, writing, we updated our working paper on clustering. Lots of time working on single formulation that captures intermediate cases, simple after-the-fact, math matches intuition. arxiv.org/abs/1710.02926
Sven Resnjanskij retweeted
My 2009 paper with @MiriamBruhn1 showed advantages of matched paired randomization. In today's post, why I am now more cautious about using this approach and like matched quadruplets instead. Thanks to @CdeChaisemartin & Yuehao Bai for discussion. blogs.worldbank.org/impactev…
Sven Resnjanskij retweeted
I've been in touch with several German universities which have expressed interest in admitting students displaced from Ukraine. If you're a displaced student, you can register your interest via this form, and I will share it with the universities: forms.gle/S2f4QURoVJWaM1PFA
Sven Resnjanskij retweeted
But with moderate trends they can differ, with logit or probit being better across different DGPs. With strong trends (not violation of PT, BTW), the LPM can even give the wrong sigh. Logit, probit never do -- so far.
Sven Resnjanskij retweeted
Sven: I've been doing a lot of simulations in the binary case, in the context of my presentation here:
stata.com/symposiums/economi…
If the trend in the control state is not very strong, logit, probit, LPM give similar answers.
Sven Resnjanskij retweeted
Am I wrong, assuming that in the binary-response DiD case, it would be an option to still run a linear DiD spec., when it'S feasable to express all RHS variables in terms of binary variables as well? My intuition is, that (except SE) OLS would give similar estimates.
Sven Resnjanskij retweeted
It is a very strong assumption. And it's especially strong when the linear form is applied to, say, a binary response. So those who say "just use OLS" seem not to realize that a linear model can be very poor in these situations. So I want to try it a few different ways.
Sven Resnjanskij retweeted
How could we know the correct transform in which parallel trends hold? I think PT is a *far* stronger assumption than most users recognize. This is relevant: jonathandroth.github.io/asse…
Sven Resnjanskij retweeted
Hot take: 2022 will be the year of backlash against Diff-in-Diff models. Users will finally realize parallel trends is often an implausible, structural assumption. E.g., What stops control group agents from modifying behavior based on observing the treated group? #EconTwitter
Sven Resnjanskij retweeted
As another analogy, we now know from @TymonSloczynski's nice REStat paper on regression adjustment that OLS of Y on 1, D, X can produce badly biased estimators of ATE and ATT. But full RA, Y on 1, D, X, D*(X - Xbar) can work very well (with overlap).
Sven Resnjanskij retweeted
We don't conclude that something is "wrong" with the OLS estimator. Rather, something is wrong with the functional form of the underlying model: we need to allow the TEs to vary with X.
OLS can be lousy when applied to a bad model and good when applied to a good model.
Sven Resnjanskij retweeted
Have I mentioned that OLS is an estimation method, not a "model"?
Fortunately, I have a firm grasp of the obvious.
Sven Resnjanskij retweeted
Really interesting DiD session at ASSA! I now realize I'm caught up through about June 2021.
There wasn't time for me to make a point about a theme running throughout the session, and that is picking on so called two-way fixed effects.
@bruno_ferman @CdeChaisemartin @jondr44
Sven Resnjanskij retweeted
Welchen Einfluss hat #Kultur – und hier im Speziellen: #Individualismus – auf kognitive #Fähigkeiten, #Lohn und andere Arbeitsmarktindikatoren? Eine neue Studie mit Daten aus 22 Ländern liefert erstmals Antworten: oekonomenstimme.org/artikel/…
