Big data and machine learning for price prediction in residential real estate – Opportunities at University of Aberdeen – Aberdeen, Scotland

Job Overview

Big data and machine learning for price prediction in residential real estate

Details

The price of residential property is not only affected by hard characteristics such as its size and location, but also by softer characteristics such as its design and floorplan. Unfortunately, most of these softer characteristics are usually not recorded in standard data sets of house transactions. Property listings on platforms have descriptions that can provide information on such softer characteristics. For instance, listings usually describe the property and provide photos of most rooms. The listing platform can be used to find out which listed properties are most visited, thus indicating what market participants are looking for in the market. A data set with most of these variables will be available for the project.

The PhD project should apply machine learning and develop automated valuation models that generate reliable price predictions for properties that are currently on the market. It is expected that the successful candidate also works with data providers.

Applicants interested in this research project should submit a more detailed research proposal (of a maximum of 2000 words) that develops some ideas and shows understanding of the existing literature based on the outline above.

Informal inquiries can be made to Dr Martin Wersing with a copy of your curriculum vitae and cover letter indicating your interest in the project and why you wish to undertake it.

The successful applicant is expected to have (or be close to graduating with) an MSc in Economics/Econometrics or in a related area, e.g., a MSc in Statistics or Computer Science with prior knowledge of prediction. Applicants should have a strong interest or experience in data preparation, machine learning, and in the implementation of projects. The project requires excellent skills in programming (R, Matlab, or Python). General knowledge in urban and housing economics is desirable.