This article presents a multiple linear regression method with mixed effects that was developed to value residential properties in the Montana state tax cycle. The approach considers the contribution of property characteristics and geographic submarket areas (SMAs). The research confirms that automated creation of model classes may yield better results than currently used methods, and that they outperform standard methods when the individual property characteristics are missing or not available. This research also confirms that the linear mixed model increases efficiency when sufficient data is available.
Multiple regression analysis
The material in this article was first presented during the 19th Annual GIS/CAMA Technologies Conference, March 2-4, 2015, in Oklahoma City, Oklahoma. The conference is jointly sponsored by the International Association of Assessing Officers (IAAO) and the Urban and Regional Information Systems Association (URISA).
Reddy, S. (2015). Residential property value estimation via linear mixed model methods. Journal of Property Tax Assessment & Administration, 12(2), 73-93. Retrieved from https://researchexchange.iaao.org/jptaa/vol12/iss2/5