Determinants of households´ consumption in Portugal - a machine learning approach

• Autor(a) principal: Noro, Catarina Vieira
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/121884
• topic: Machine learning algorithms; Feature selection; Lasso regression; Elastic; Domínio/Área Científica::Ciências Sociais::Economia e Gestão

Machine Learning has been widely adopted by researchers in several academic fields.Although at a slow pace, the field of economics has also started to acknowledge the pos-sibilities of these algorithm based methods for complementing or even replace traditionalEconometric approaches. This research aims to apply Machine Learning data-driven variable selection models for accessing the determinants of Portuguese households’ consumption using the Household Finance and Consumption Survey. I found that LASSO Regression and Elastic Net have the best performance in this setting and that wealth related variables have the highest impact on households’ consumption levels, followed by income, household’s characteristics and debt and consumption credit.