A reader asked for my opinion on this article on missing data mdpi.com/1999-4893/16/1…. The authors do not relate their method to previous works in missing data (eg ucla.in/3r87JMZ) and, so, it's very hard to judge. Moreover, we know that it is impossible to have an…


Even Sir David agreed that 'missing at random' was a a disingenuous ploy by statisticians when we discussed his methods for survival analysis. Unfortunately, competing risks are not present in #CausalAnalysis either

soboleffspaces's tweet image. Even Sir David agreed that 'missing at random' was a a disingenuous ploy by statisticians when we discussed his methods for survival analysis. Unfortunately, competing risks are not present in #CausalAnalysis either

#causalanalysis treats every variable according to its position in the graph. Is there anything special that should be done with "competing risks"?


Well, it's not entirely clear how to position competing risks on the graph. Some statisticians place a competing event as a mediator between exposure and the primary outcome of interest. They may not realize that causal mediation analysis addresses a different question: "Would…


United States Trends
Loading...

Something went wrong.


Something went wrong.