Describe the bug
The populationsim/configs/controls.csv was created with input from the E&F, ABM, and Econ teams. Somewhere along the way, total population at the MGRA-level was de-prioritized and both total households and labor force participants by race/ethnicity were given higher importance weights than the MGRA-level sex controls, which function as a surrogate for a total population control.
To Reproduce
See the populationsim/configs/controls.csv file.
Expected behavior
A synthetic population program being used to feed into an Activity Based Model should first and foremost match total population at the MGRA-level above all other controls, then households. All other controls, while important, pale in comparison to total population as Activity Based Models operate at the person level while taking into account person/household attributes.
Potential resolution
Testing will be needed but the MGRA-level sex controls should be set to have at least as much importance as total households, if not more. Labor force participants by race should never supersede those controls.
Additional Context
This remains icebox-ed until all 2025 Regional Plan modeling is complete as major changes to controls that were deliberated and decided upon early in the process should not be altered at this point in time.
Describe the bug
The
populationsim/configs/controls.csvwas created with input from the E&F, ABM, and Econ teams. Somewhere along the way, total population at the MGRA-level was de-prioritized and both total households and labor force participants by race/ethnicity were given higher importance weights than the MGRA-level sex controls, which function as a surrogate for a total population control.To Reproduce
See the
populationsim/configs/controls.csvfile.Expected behavior
A synthetic population program being used to feed into an Activity Based Model should first and foremost match total population at the MGRA-level above all other controls, then households. All other controls, while important, pale in comparison to total population as Activity Based Models operate at the person level while taking into account person/household attributes.
Potential resolution
Testing will be needed but the MGRA-level sex controls should be set to have at least as much importance as total households, if not more. Labor force participants by race should never supersede those controls.
Additional Context
This remains icebox-ed until all 2025 Regional Plan modeling is complete as major changes to controls that were deliberated and decided upon early in the process should not be altered at this point in time.