Chicago Model

Data inputs (available at this link)

acs_income.csv

• geoid: tract FIPS code covering all counties in Chicago MSA
• variable: Census variable ID
• income_bin: household income in the past 12 months (in 2020 inflation-adjusted dollars)
• share: share of income bin
• estimate: number of households at income bin level

tract_firstam.csv

• tract_fips: tract FIPS code covering all counties in Chicago MSA
• property_class: one of the following categories 'Residential', 'Exempt', 'Commercial', 'Vacant', 'Industrial', 'Office', 'Agricultural', 'Recreational', 'Missing', 'Transportation'
• buildingarea: building square footage that can most accurately be used for assessments
• lotsizesqft: total area measurement of the land in square feet
• market_totalvalue: market value of land and building value (adjusted from assdtotalvalue using assessment level of 10% for Cook County Residential, 25% for Cook County Commercial, and 33.33% for all other properties statewide)

tract_ttm.csv (generated using r5r)

• from_id: tract FIPS code in Cook County
• to_id: tract FIPS code in Cook County
• travel_time_p50: travel time (in minutes) from the centroid of the "from" tract to the centroid of the "to" tract. (Note: if the centroid is not within the shape of the tract, then we use a "point on surface" which approximates a center point that is within the tract boundaries). p50 indicates that the result is the 50 percentile or median travel time. For more on this, see the documentation of r5r travel_time_matrix()).

Notes:

• Rethinking Detroit replication code link.
• Overview of data sources at this link.

Next steps:

• Use lehdr to process LODES data.