This is the repository for the English version of the Future Elderly Model, originally developed at the Leonard D. Schaeffer Center for Health Policy and Microsimulation.
MIT License
3
stars
1
forks
source link
Implement predictive mean matching for imputation of missing data. #8
Hotdeck imputation is used for imputing some important variables (Education, drinkd_e). We chose to use hotdecking initially as it maintains the distribution found in the original data, and imputes actual true values from similar records that do no have missing data. It has been brought to my attention that Predictive Mean Matching (PMM) would probably be a better alternative to hotdeck imputation, so I should replace hotdecking in the population generation scripts with PMM.
Hotdeck imputation is used for imputing some important variables (Education, drinkd_e). We chose to use hotdecking initially as it maintains the distribution found in the original data, and imputes actual true values from similar records that do no have missing data. It has been brought to my attention that Predictive Mean Matching (PMM) would probably be a better alternative to hotdeck imputation, so I should replace hotdecking in the population generation scripts with PMM.