Publicação

Sentiment analysis-based prediction of -meme stock- price movements

Ver documento

Detalhes bibliográficos
Resumo:This paper investigates the relation of daily twitter sentiment to the stock prices of two portfolios for a three-year time-period, one consisting of popular meme stocks, the other of NASDAQ stocks. The objective is to detect differences in the behavior of both portfolios in this relation. In its sentiment classification, the paper makes use of Textblob, customized VADER, roBERTa as well as a manually trained GRU neural network while relying on classic binary classifiers to predict price movements. The results indicate a stronger predictability for the NASDAQ portfolio, most likely due to issues of information quality.
Autores principais:Busch, Benedikt
Assunto:Sentiment Aanalysis Meme stocks Twitter
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 paper investigates the relation of daily twitter sentiment to the stock prices of two portfolios for a three-year time-period, one consisting of popular meme stocks, the other of NASDAQ stocks. The objective is to detect differences in the behavior of both portfolios in this relation. In its sentiment classification, the paper makes use of Textblob, customized VADER, roBERTa as well as a manually trained GRU neural network while relying on classic binary classifiers to predict price movements. The results indicate a stronger predictability for the NASDAQ portfolio, most likely due to issues of information quality.