Does (1) include partially observed/non-identifiable situations/P>>N too? Or just identified but small N?
In my mind, the two most important methodological issues for applied policy research are (1) inference with smallish N and (2) estimating long-run effects with short-run data.
-
-
-
In my line of work, the latter is the more typical challenge (just identified, but small N).
Kraj razgovora
Novi razgovor -
-
-
We should push for more Bayesian inference then?
-
That's one approach, and it allows for formal synthesis of various types of evidence (e.g., http://www.macartan.nyc/projects/integrated-inferences/ …). In a frequentist setting, there are also possibilities -- e.g., multivariate inference: https://www.journals.uchicago.edu/doi/abs/10.1086/689287 …
- Još 1 odgovor
Novi razgovor -
-
-
Applies equally to industry.
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.