PolicyEngine / synthimpute

Python package for data synthesis and imputation using parametric and nonparametric methods, and evaluation of these methods.
MIT License
11 stars 6 forks source link

Target a total in rf_impute #40

Closed MaxGhenis closed 3 years ago

MaxGhenis commented 3 years ago

rf_impute should take an argument for a target total (e.g. an administrative total) to hit. It can achieve this target by adjusting the quantiles to select. Currently, it samples randomly from the conditional distribution, effectively using the uniform distribution to select quantiles. To achieve the target while still selecting realistic values (those in the conditional distribution), it can instead adjust the distribution of quantiles to select, by using something other than the uniform distribution.

MaxGhenis commented 3 years ago

Fixed in #45