Q1. Which function could be used to do the MCAR test?

Q2. Which code could be used to impute 10 datasets using multiple imputation technique?

Q3. How could you extract all the 10 imputed datasets into one complete daatset from Q2?

Q4. You have datasets 'imputation' from Q2 and 'impdata' from Q3. Which code can be used for variable selection using the Stack approach?

A. fit4 <- with(imputation, expr), where expr includes unevaluated expressions of the initial model and the model chosen by the stepwise algorithm.
B. fit4 <- step(fit0, scope = scope0, trace = FALSE), where fit0 is the inital model and scope0 includes the range of models examined in the stepwise search on the dataset named 'imputation'.
C. fit4 <- step(fit0, scope = scope0, trace = FALSE), where fit0 is the inital model and scope0 includes the range of models examined in the stepwise search on the dataset named 'impdata'.
D. fit4 <- D1(fit0, fit1), where fit0 and fit1 are two models for comparison fitted on the dataset named 'imputation'.

Q5. Which function can be used to combine estimates from imputed datasets by pooling rules?