Login
iscover
Researchers at the University of South Australia (UniSA) have developed a new technique they say will help the property sector gain a more realistic, reliable and practical view of the value of property.
The technique has been made and tested using over 30 years of historical sale information in metro Adelaide.
It uses purpose-built machine-learning algorithms to process huge amounts of data about housing, urban structure, and amenities. This makes it possible to measure the effects of urban planning policies on housing value.
UniSA urban planning specialist and lead researcher Dr Ali Soltani said the method has implications for the real estate, infrastructure, and urban planning industries.
“This research has implications for policymakers by providing insights into the potential impacts of urban planning — such as infill regeneration, master-planned communities, gentrification, and population displacement — and infrastructure provision policies on the housing market and subsequent local and regional economy,” he said.
“By taking into account how complex infrastructure factors like road and public transportation networks, commercial centres, and natural landscapes affect the value of a home, our model is especially useful for improving the accuracy of current land value predictions and lowering the risks of traditional property valuation methods, which are based on human experience and limited data.”
The researcher also explained that the model, made with Professor Chris Pettit from UNSW’s City Futures Research Centre, could be expanded to include other macro and microeconomic factors, such as changes in interest rates, employment rates, and the effect of COVID-19, by using big data technologies.
“This model has the potential to be used as a decision-support platform for a variety of stakeholders, including home buyers and sellers, banks and financial agents, investors, the government, and insurance or loan agents,” Dr Soltani continued.
“Our technique makes it simpler for stakeholders and the general public to apply the findings of sophisticated models on historical or real-time data from multiple sources, which have previously been almost black-box and expert-oriented.”
Newsletter
