1	for j ← 1 to q do
2	Impute the missing values of Xj.
3	end
4	for m ← 1 to M do
5	repeat
6	for j ← 1 to q do
7	Fit the imputation model for Xj using the samples for which Xj are observed and impute the missing values of Xj by iMI, avgmMI, cslMI, or siMI;
8	end
9	until multiple times;
10	Fit the analysis model and obtain ${\widehat{\mathit{\theta }}}_m$ and $\mathrm{Cov}\left({\widehat{\mathit{\theta }}}_m\right)$;
11	end
12	Combine the results by Rubin’s rule to obtain $\widehat{\mathit{\theta }}$ and $\mathrm{Cov}\left(\widehat{\mathit{\theta }}\right)$;