Close
Economic analysis of the Real Estate market with statistical machine learning and big data
School Research Theme: Economics and finance of the built environment
Research supervisors: Dr Stanimira Milcheva
Start date: October 2017
Considering our role in the housing market, we might be a property owner, buyer, seller, tenant, or investor. Housing is closely related to our daily life. We can’t talk about it without talking about its value, the price, and it is indeed often one of the most significant components of household wealth. Unfortunately, we can only have a rough estimation based on online search, recent resale in neighborhood or property agents. But this won’t fit the need for risk management when our decision-making involves property purchase. What will be the transaction price? How long will it take to reach an agreement and finish the transaction? They are the two key tangled questions we need to figure out before making the decision. In the past 40 years’ housing economics, researchers have done a large number of studies on this price and Time-on-market trade off without reaching an agreement.
I developed an up-to-date database integrated with multiple sources of property information, which is one of the largest in this kind of studies. Then aiming to have a deeper understanding of the trade-off and provide a more accurate prediction of the answers of the two key questions, I utilize modern statistical and machine learning methods, industry grade workflow and best practices while developing quantitative academic research methods.