I’m interested in your take. I’ve used PS scoring in several studies and haven’t seen much added benefit/utility.
-
-
-
Nothing can match a well-done RCT. Don't like PS matching, you lose subjects, lose power while matching. IPW is way better, gets u closest to RCT, still conditional on known confounders. Read chptrs 3 & 12 of Hernan's Causal inference book https://www.hsph.harvard.edu/miguel-hernan/causal-inference-book/ …
#FOAMed - Još 3 druga odgovora
Novi razgovor -
-
-
Agree, propensity scores are a great tool. As elaborated on only yesterday
@FustoloSusanna
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.