Developing a predictive property valuation model for residential property. Case study-Kawempe division

Abstract

The property market is one of the greatest evolving sectors in the universe apart from food to a human life. These property market should meet the criteria of adequacy, affordability and decent way of performing transactions. As these market evolves, in relation to technology advancements, there is need for a proper way of keeping property values to check on the growth and how to improve these sector especially for the planning and growth of country. The property market is interwoven with property valuations which is done for a number of purposes specially to process for mortgage financing, taxation and compensation. In Uganda, the property valuation system is still manualized and this system greatly has loopholes in relation to the global advancements, it is impossible to verify all the information provided by the valuers and at the same time, the process is lengthy and time consuming and involves repeated works and the planning authorities are neither efficient nor transparent. The research focused on developing a predictive model that not only stores data values but also quickens the process of valuation in a simpler manner. This model was subject to residential property and can be adopted to any type of property. The study employed a case study of Kawempe division. Descriptive designs and modeling techniques on a sample 120 properties that were selected by obtaining property values carried in the last 5 years that were got from various valuation firms was used. This data was modeled using a hybrid software called anaconda and tested using two algorithms that is to say; Ridge Linear Regression and Multiple Linear Regression to see which one performs best. The study has revealed since the real estate sector has become one of the safe havens for investors seeking meaningful returns, there is a great need for the valuation systems in Uganda to adopt artificial intelligence as the economy evolves. The researcher put out several recommendations that could be a solution to the continuous growth of the property sector and as technology advances, machine learning is vital because this technology focuses on the development of computer programs that can access data and use it learn for themselves.