Financial time series forecast using artificial neural networks: a comparative in the 2008 crisis
Artificial Neural Networks (ANN) have been used in different segments inside the area of finance such as stock prices and market indices forecast. This article seeks to measure the power of ANN on the Bovespa Index and the prediction of stock prices, verifying their forecast power even in times of c...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universidade Estadual Paulista
2018-03-01
|
| Series: | GEPROS: Gestão da Produção, Operações e Sistemas |
| Subjects: | |
| Online Access: | http://revista.feb.unesp.br/index.php/gepros/article/view/2016 |
| Summary: | Artificial Neural Networks (ANN) have been used in different segments inside the area of finance such as stock prices and market indices forecast. This article seeks to measure the power of ANN on the Bovespa Index and the prediction of stock prices, verifying their forecast power even in times of crisis. Therefore, time series of over a decade were extracted from Yahoo! Finance, including the period of the subprime crisis and its temporal neighborhoods. ANN were performed using Matlab 2016a software with satisfactory results, which were evaluated by scattergrams errors and Mean Absolute Percentage Error (MAPE) method. |
|---|---|
| ISSN: | 1984-2430 |