Stock trading and prediction using multi-layer perceptron neural networks
Stock price prediction has always been a choice problem to solve for stock enthusiast and investors alike. Everyone would like to remove the shade of uncertainty over the stock’s future prices and trends. Tackling this problem with neural networks has been done by many for decades. This project appl...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
2016
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/67849 |
| Summary: | Stock price prediction has always been a choice problem to solve for stock enthusiast and investors alike. Everyone would like to remove the shade of uncertainty over the stock’s future prices and trends. Tackling this problem with neural networks has been done by many for decades. This project applies a multi-layer perceptron model with a moving window simulation to the stock price prediction problem, based on a paper written by Turchenko et al. Various experiments were carried out to determine the parameters of a better model with higher accuracy. Comparisons on the influence of each parameter over the results were done in later parts of the report. |
|---|