Determinants of Residential Property Value in Nigeria – A Neural Network Approach

This study investigated, by means of artificial intelligent system, theinfluence of residential real estate property characteristics on propertyvalues (prices) in Nigeria, using two major cities (Benin and Lagos) asexamples. It revealed a high positive linear correlation between propertycharacteristics and the property market values; an indication that thesecharacteristics reasonably predict property market values. The studydemonstrated that although several property characteristics can beidentified with residential real estate properties, only are few importantones have significant impact on the market values of such properties. Itidentified nine (9) property characteristics that have relatively strongimpact on market values (prices) and to that extent influence the salesand purchase decisions of sellers and buyers in Nigeria. The results ofthe study should enable Real Estate Professionals to make fair estimatesof the market values of residential real estate properties given the features/characteristics of such housing units. This would aid rapidvaluation, help to improve housing quality and make possible massevaluation of properties. The study recommends to real estatepractitioners and other professionals, amongst others, to use theknowledge of significant property features/characteristics for moreefficient valuation, improved quality of their sales/purchase decisionsand proper management of residential housing units.