Stock returns and Google Search volume data – an analysis on the Portuguese and American market

• Autor(a) principal: Streicher, Annette Rachel
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/70650
• topic: Investor attention; Investor sentiment; Forecasting returns; Google SVI; Domínio/Área Científica::Ciências Sociais::Economia e Gestão

This paper examines the forecasting power of Google Search Volume Data on market returns in the light of Behavioral Finance. The research is twofold: we investigate the ability of investor attention as well as investor sentiment to predict future returns. We consider weekly time series data from 2008 to 2018 for two American market indices and the Portuguese market. Investor attention is captured by search volume of the index’s names, i.e. DJIA, S&P500 and PSI20. Investor sentiment is simulated robustly by constructing two modified sentiment indices. We apply VAR models and Granger Causality and show that our proxies for investor attention do not provide significant forecasting information opposite to previous research. Similarly, investor sentiment indices constructed with English searched terms cannot predict market returns. However, both investor sentiment indices constructed with Portuguese words reveal significant precedence.