ABSTRACT
The national government plays a negligible role in property tax administration, allowing each state government to design its own property tax system. The California State Board of Equalization first used multiple regression analysis operationally in property tax administration to value single-family houses in about 1968, although its usefulness in property valuation was recognized in 1919. This chapter discusses specific modelling considerations that apply to various types of residential property. Models for multi-family residential properties can take several forms. Property characteristics often important in such models include location, average unit size, average baths per unit, construction grade, age and condition, parking, common area facilities, and land size or land/building ratio. Computer assisted mass appraisal (CAMA) system includes a flexible database management system, a market model module, a comparable sales selection module, an income capitalization module, a cost module, and a sales ratio analysis module. Objectives of CAMA models include predictability, explainability, and stability.
