Publicação

Leveraging google search queries to help predict house prices in Portugal

Ver documento

Detalhes bibliográficos
Resumo:This work project contributes to the current literature on using Google search queries to predict economic activity. We demonstrate, using the two-step Error-Correction Model (ECM) by Engle and Granger (1987), that specific search queries, also known as Google Trends, are related to house prices in Portugal. For out-of-sample forecasts, our ECM model with the Google Trends variables performed significantly better predicting one year ahead, in which, the Mean Absolute Error was reduced by over 30% compared to our baseline model. Until now, conventional economics has not leveraged this highly accessible digital data in their models, we hope this will change.
Autores principais:Sistovaris, Nicholas
Assunto:Econometrics Housing Google trends Forecasting Error-correction model
Ano:2023
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Universidade Nova de Lisboa
Idioma:inglês
Origem:Repositório Institucional da UNL
Descrição
Resumo:This work project contributes to the current literature on using Google search queries to predict economic activity. We demonstrate, using the two-step Error-Correction Model (ECM) by Engle and Granger (1987), that specific search queries, also known as Google Trends, are related to house prices in Portugal. For out-of-sample forecasts, our ECM model with the Google Trends variables performed significantly better predicting one year ahead, in which, the Mean Absolute Error was reduced by over 30% compared to our baseline model. Until now, conventional economics has not leveraged this highly accessible digital data in their models, we hope this will change.