The process of MI is as following:
In the first step, the dataset with missing values (i.e. incomplete dataset) is copied several times. Then in the next step, the missing values are replaced with imputed values in each copy of the dataset. In each copy, slightly different values are imputed due to random variation. This results in multiple imputed datasets. In the third step, the imputed datasets are each analyzed and the study results are then pooled into the final study result.
My question is that in the third step, after we analyze each dataset and get coefficients in the regression function in each dataset. In the end, how can we deal with these coefficents? Do we just to calculate the average?
It is usually achieved by pooling the results according to Rubin's Rules - see https://bookdown.org/mwheymans/bookmi/rubins-rules.html
Kelvyn Jones provided a very useful link. Some software programs (e.g., Mplus) will automate the pooling process and provide you with a summary output of the pooled estimates and their standard errors so that you won't even have to deal with the individual coefficients or how to pool them properly.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression