A carregar...
Publicação
Predicting how Macroeconomic Factors can impact the Housing Market in Portugal Using Machine learning Techniques
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 2.83 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
Background: The Portuguese Housing Market has been growing in the last few years as OECD studies mention. However, there has been a lack of scientific research addressing this phenomenon in Portugal.
Objectives: This study seeks to fill this gap and to understand how Macroeconomic factors can impact the Portuguese housing market.
Methodology: In this study, the Vector Autoregression model (VAR) and Gated Recurrent Units model were applied, using two different variable approaches. One utilizing a dimensionality reduction technique, factor analysis, and the other using the original uncorrelated variables.
Results: The results of this study conclude that utilizing the Mean Squared Error that the VAR model using the factor analysis dataset outperforms the other VAR model and the GRU models, having an MSE of 0.11. In this model, the dependent variable has a positive relationship with lagged one-time economy and a negative relationship with lagged one-time nominal house prices and lagged four-times nominal house prices. The economy factor has high factor loadings with GDP, Euribor 3M (interest rate), and Tourism and there exists scientific research in other countries associating these three variables with the housing prices.
Conclusion: This study concludes that the macroeconomic variables impact the Portuguese housing market specifically economic-centric variables such as GDP, interest rates and, tourism.
Descrição
Dissertation presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics, specialization in Data Science
Palavras-chave
Housing Market Macroeconomic Variables Vector Autoregression Model Gated Recurrent Units SDG 8 - Decent work and economic growth