Twitter | Search | |
Tweets
Paul Hünermund 6h
Replying to @PHuenermund
Reply Retweet Like
Paul Hünermund 6h
I think it's totally fine that we still teach SPSS to our students in the master track. It best prepares them for a career in the private sector.
Reply Retweet Like
Paul Hünermund 7h
Replying to @nickchk
Biasing paths always start with the treatment (or the IV in this case) and end with the outcome Y. Therefore biasing paths cannot have Y as an intermediate variable, otherwise the graph wouldn't be acyclic. So the fact that Y is a collider shouldn't play a role here, I think.
Reply Retweet Like
Paul Hünermund 16h
War schon immer meine Rede, dass AKK gar nicht geht. Aber das Counterfactual Merz kennen wir halt auch nicht.
Reply Retweet Like
Paul Hünermund 19h
Replying to @AdeGendre
Although I agree that francophonie should be corrleated with distance to the French border, and regions closer to France might be affected more by inner-French affairs. I don't know how the authors deal with this.
Reply Retweet Like
Paul Hünermund 19h
Replying to @AdeGendre
As far as I understant they look at francophone regions outside of France, which had easier access to French ideas but you should not have been affected as much by internal politics.
Reply Retweet Like
Paul Hünermund 19h
Replying to @Mohdfaour89
I think it's very unlikely that I'll be invited to review this paper. 😉
Reply Retweet Like
Paul Hünermund 19h
Replying to @HolgerSteinmetz
I think this is because traditional SEM is purely based on linear models and people associate it with such. Abother reason is that the SCM terminology uses assignment operators instead of equations to capture the asymmetric nature of causal relationships.
Reply Retweet Like
Paul Hünermund 20h
Replying to @PHuenermund
I wonder whether the standard errors are robust to spatial autocorrelation though (just kidding, didn't have a closer look at the paper)? 😉
Reply Retweet Like
Paul Hünermund 20h
Interesting new study on how new secular cultural norms spread from France throughout Europe in the 19th centurity and reduced fertility rates in the affected regions.
Reply Retweet Like
Paul Hünermund retweeted
Carlos Serrano Jun 16
On my way to Munich to present my new work with G. Chondrakis and R. Ziedonis on information disclosure and the market for technology acquisitions at the Munich Summer Institute
Reply Retweet Like
Paul Hünermund Jun 16
I will probably never understand why this kind of reactionary BS is just not simply forbidden in the U.S.?
Reply Retweet Like
Paul Hünermund retweeted
Rudi Bachmann Jun 15
German progressives (economists) and others: if you haven’t done so, please read this essay by Daron Acemoglu. This modern social democratic program gets you above 30 pc, at least:
Reply Retweet Like
Paul Hünermund retweeted
Paul Hünermund Jun 15
There's always the possibility to apply this neat little trick. 😉
Reply Retweet Like
Paul Hünermund Jun 15
There's always the possibility to apply this neat little trick. 😉
Reply Retweet Like
Paul Hünermund Jun 14
Da setzt ihr auf das falsche Pferd, glaube ich. Weniger konservativ eingestellte Wähler werden die Gefahr einer Austria-Koalition weit weniger diskontieren als ihr denkt.
Reply Retweet Like
Paul Hünermund Jun 14
Reply Retweet Like
Paul Hünermund Jun 14
Der gleiche Mismatch zwischen Wähler- und Parteipräferenzen gibt es auch bei der CDU. Bin mir nicht so sicher, wie die Entscheidung lauten würde, wenn man zwischen AfD und Grünen wählen müsste.
Reply Retweet Like
Paul Hünermund Jun 14
Replying to @schnellenbachj
Nee falsch, die Stimme für Grün macht 1. wahrscheinlicher und 4. unwahrscheinlicher, gerade unter einem Kanzlerkandidaten AKK oder Merz.
Reply Retweet Like
Paul Hünermund Jun 14
Which is a real pity if you have Chinese students. They usually make nice gifts. 😉
Reply Retweet Like