Overview

A common problem in generating a representative synthetic population is that not all attributes of interest are present in the sample. The purpose is to enrich the synthetic population with additional attributes, after the synthetic population is generated from the original sample.

In many cases, practitioners only have access to aggregated data for important socio-demographic attributes, such as income, education level.

Bhepop2 treats the problem to enrich an initial synthetic population from an aggregated data provided in the form of a distribution like deciles or quartiles.

See a detailed description of the algorithm inputs in the Inputs section

For an explanation of the algorithm logic, see the Algorithm section.

You can find a detailed documentation of the code in the API reference.

More information about Bhepop2 here.