ABSTRACT
The recursive model consists of two equations. The first explains the average number of persons per household in the neighborhood as a negative function of the median age of the residents and as a positive function of the owner-occupancy rate expressed as a percentage. The second equation explains vehicular density, or vehicles per household, as a positive function of average household size and per capita income. The model was estimated by using Ordinary Least Squares on data from 38 census blocks in San Leandro. This model assumes that the average household size of occupied accessory apartments would be 1.25. If both the apartments and the persons occupying them were included in the calculation, average household size in the neighborhood would appear to decline as the percentage of houses with conversions increases. Projected vehicles per lot was computed using the parameters of the regression equation and then adding the additional cars per lot generated by the assumed number of accessory conversions.
