The Impact of COVID-19 on Perth House Price

A Machine Learning Perspective

DOI

10.2991/aebmr.k.220307.203How to use a DOI?

Keywords

Catboost; Evaluation Method; Machine Learning; Risk Analysis

Abstract

After analyzing suburb information, school-area concentration, subway stations distribution, and 20 other parameters, a machine learning project using Catboost regression is utilized to predict house prices in Perth. To improve the prediction accuracy, a few innovative variables are created, like “Average_area” which indicates the average floor area per bedroom. Simultaneously, several regression algorithms were selected to apply to the model, and Catboost is selected based on a newly designed evaluation method in this research. With the project, using an alternative approach of the Difference-in-Difference (DID) method allows the product to keep the control group as the pre-COVID terms to analyze the impact made by the pandemic to house price market in Perth. The results show that there is no evidence indicating the house prices have been impacted by the pandemic. However, it could be noticed that this result only works for a city like Perth which does not have many cases and a statement of general lockdown does not have an impact on the real estate market. A few evaluation methods are utilized to make the judgment and further cater industry’s needs and better filter out the models suitable for business. These methods are expected to apply in the industry for trade-in evaluation and future market forecasting.

Copyright

© 2022 The Authors. Published by Atlantis Press International B.V.

Open Access

This is an open access article under the CC BY-NC license.

Download article (PDF)

TY  - CONF
AU  - Yue Hu
PY  - 2022
DA  - 2022/03/26
TI  - The Impact of COVID-19 on Perth House Price
BT  - Proceedings of the 2022 7th International Conference on Financial Innovation and Economic Development (ICFIED 2022)
PB  - Atlantis Press
SP  - 1225
EP  - 1232
SN  - 2352-5428
UR  - https://doi.org/10.2991/aebmr.k.220307.203
DO  - 10.2991/aebmr.k.220307.203
ID  - Hu2022
ER  -