Machine Learning-Based Approaches to Forecasting Housing Prices in the Canadian Market

Yuhao Wen

2025

Abstract

Housing prices are one of the key indicators for measuring the state of the real estate market. Establishing an efficient housing price forecasting model is of considerable importance to consumers, investors, and policymakers. In the context of Canadian research, scholars have constructed several regression models to predict housing prices. However, there remains a paucity of systematic research on model comparison and hybrid models. This study utilizes Canadian housing price data and employs a data preprocessing technique involving least absolute shrinkage and selection operator (LASSO) feature selection. Multiple regression models are then constructed including multiple linear regression (MLR), random forest (RF), extreme gradient boosting (XGBoost), and a hybrid model integrating RF and XGBoost. During the model building process, GridSearchCV method is applied to perform hyperparameter tuning for the machine learning models RF and XGBoost. The models are subsequently compared and analyzed using metrics including Mean Absolute Error (MAE), Root Mean Squared Error (RMSE) and R-squared to identify the most effective model for housing price prediction within the Canadian context. The study revealed that the hybrid model, comprising a linear weighted combination of RF and XGBoost, demonstrated the most efficacy in housing price forecasting.

Paper Citation

in Harvard Style

Wen Y. (2025). Machine Learning-Based Approaches to Forecasting Housing Prices in the Canadian Market. In Proceedings of the 2nd International Conference on Data Science and Engineering - Volume 1: ICDSE; ISBN 978-989-758-765-8, SciTePress, pages 348-354. DOI: 10.5220/0013689900004670

in Bibtex Style

@conference{icdse25,
author={Yuhao Wen},
title={Machine Learning-Based Approaches to Forecasting Housing Prices in the Canadian Market},
booktitle={Proceedings of the 2nd International Conference on Data Science and Engineering - Volume 1: ICDSE},
year={2025},
pages={348-354},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013689900004670},
isbn={978-989-758-765-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 2nd International Conference on Data Science and Engineering - Volume 1: ICDSE
TI - Machine Learning-Based Approaches to Forecasting Housing Prices in the Canadian Market
SN - 978-989-758-765-8
AU - Wen Y.
PY - 2025
SP - 348
EP - 354
DO - 10.5220/0013689900004670
PB - SciTePress