I have simple question after running multiple imputation what the purpose of pooling?

Suppose if i run a multiple imputation using method cart , after running this imputation technique i get very stable density plot between observed and imputation values. So my understanding is that after that i can assign any imputation to the data which are missing as my density plot telling me imputation process is fine for all m=20

imp<- mice(data.combined_subset,m=20,maxit=50,method="cart",seed=147) densityplot(imp, xlim = c(2.5, 80), ylim = c(0, 0.1)) data.combined_subset$age<-complete(imp,1)

After that what is the purpose of pooling?

model1 <- with(imp,lm(Age~ X1+X2+X3))



What is pooling usage in imputation? Do we really get some value from pooling and impute into data.combine_subset$Age which has null value?




Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.