Also, @edrubin's materials on instrumental variables (and all things econometrics) are magisterial https://raw.githack.com/edrubin/EC421W19/master/LectureNotes/11InstrumentalVariables/11_instrumental_variables.html#42 … #econtwitter
-
-
Prikaži ovu nit
-
-
SUPER naive question since I'm 100% not an expert at all on this stuff. When dealing with covariates, what's the rule for which things go in which stage? For instance, in this example with Card's proximity-to-college instrument, the same controls go on both sides. Why?!pic.twitter.com/jl4Eyrdmwi
Prikaži ovu nit -
Does this happen all the time, or are there instances where controls would go only in first stage or only in second stage?
Prikaži ovu nit
Kraj razgovora
Novi razgovor -
-
-
You can't prove exclusion but you can test it no?
-
That’s why I have arrows going both ways. You can get cor(instrument, outcome) with stats, but still have to justify it with theory
Kraj razgovora
Novi razgovor -
-
-
Thanks this is very helpful!
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Nice necessary condition.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
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.