Loading...

S&P500 Intelligent Trading Using Neural Networks

Hoseinzade, Saeid | 2009

580 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 39863 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Akhavan Niaki, Taghi
  7. Abstract:
  8. This project tries to select the inputs which really affect the change in the direction of S&P500. For this purpose, design of experiments and analysis of variance are used. T tests are carried out to calculate the statistical significance of mean differences. Experiment results indicate that the designed neural networks with the selected inputs significantly outperform the traditional logit model with respect of the number of correct predictions. Moreover, real trades are simulated using the neural network predictions in the test period and the results show that using the designed neural network can significantly increase the income.

  9. Keywords:
  10. Neural Networks ; Experiments Design ; Variance Analysis ; Logit Model ; Financial Time Series

 Digital Object List

 Bookmark

No TOC