This compares the overlap between three 1-week (Monday - Friday) searches for the Charlottesville area on AirBnB.
## [1] 56
## [1] 56
## [1] 68
## [1] 68
## [1] 123
## [1] 58
## [1] 103
## Simple feature collection with 136 features and 2 fields
## geometry type: POINT
## dimension: XY
## bbox: xmin: -78.519 ymin: 38.01136 xmax: -78.45405 ymax: 38.06301
## epsg (SRID): 4326
## proj4string: +proj=longlat +datum=WGS84 +no_defs
## # A tibble: 136 x 3
## # Groups: room_id [136]
## room_id n geometry
## * <chr> <int> <POINT [°]>
## 1 11143292 3 (-78.49218 38.01799)
## 2 11144486 2 (-78.47495 38.02022)
## 3 1138005 2 (-78.46057 38.04457)
## 4 1148833 3 (-78.47035 38.04)
## 5 11584825 3 (-78.50173 38.04602)
## 6 11733216 2 (-78.51324 38.03833)
## 7 1186869 2 (-78.49236 38.04001)
## 8 12056699 2 (-78.5007 38.0213)
## 9 12172911 2 (-78.50927 38.04144)
## 10 12218955 2 (-78.51827 38.0167)
## # … with 126 more rows
This plot suggests we may not being going far enough into the future for availability.
Considering listings with 2 or more bedrooms as a proxy for houses that would otherwise be available for rental.
Very low overlap with areas zoned non-residential (all zones exlcuding R-1, R-1S, R-2, R-2U, R-3 )
A lot of restricted locations are being falsly called as “by-right” because this zoning information is derived from parcel level data, which is not continuous across streets and property lines.