MI_IMPUTE_FROM: Stata module to impute using an external imputation model

mi impute from fills in missing values using an estimated imputation model estimated in one or multiple studies. A quantitative missing variable can be imputed passing the estimates of 99 (q=0.01(.01).99) linear predictors of quantile regression models. A categorical missing variable with k levels can be imputed passing k linear predictors of multinomial logistic regression models. A binary (0/1) missing variable can be imputed passing the linear predictors of logistic regression models. The command assumes that variables used in the imputation model are available in the current data. The mi_impute_from_get is reading the files containing the estimated models, estimate a weighted inverse variance least square model to combine regression coefficients of the imputation model across studies, and formatting matrices to be passed to mi impute from.