Informática

URI Permanente desta comunidade

Centro: CT
Telefone: (27) 4009 2324 R*5126
URL do programa: http://www.informatica.ufes.br/pt-br/pos-graduacao/PPGI

Agora exibindo 1 - 1 de 1

Preditor de alto desempenho para retornos de ações baseado em redes neurais sem peso
(Universidade Federal do Espírito Santo, 2011-08-30) Almeida, André Gustavo Coelho de; Freitas, Fábio Daros de; Souza, Aberto Ferreira de; Gonçalves, Claudine Santos Badue; França, Felipe Maia Galvão
This work presents a new weightless neural network-based time series predictor that uses Virtual Generalized Random Access Memory weightless neural network, which does not store knowledge in their connections but in Random Access Memories (RAM) inside the network’s nodes, or neurons. This new predictor was evaluated in predicting future weekly returns of 46 stocks from the Brazilian stock market and compared with neural autoregressive predictors based on feedfoward neural networks trained with the backpropagation algorithm. Our results showed that weightless neural network-based predictors can produce predictions of future stock returns with the same error levels and properties of baseline autoregressive neural network predictors, however, running 5.000 times faster.