| Title | Application of decision trees and artificial neural networks in financial markets / |
|---|---|
| Translation of Title | Sprendimų medžių ir dirbtinių neuroninių tinklų pritaikymai finansų rinkose. |
| Authors | Blažiūnas, Saulius |
| Full Text |
|
| Pages | 75 |
| Keywords [eng] | dirbtinis neuroninis tinklas, sprendimų medis, C5.0, ateities sandoriai, finansinių duomenų prognozė, artificial neural network, decision tree, C5.0, futures, financial data prediction |
| Abstract [eng] | This research compares the performance of artificial neural network against C5.0 decision tree performance. The aim is to see which one is more suitable for financial data prediction and automated trading strategy development. The evaluation is performed on out of sample/testing data. 45 most liquid futures of various financial sectors are used in simulations with 30 most popular technical indicators derived from price and volume data. Conclusions are made from 16,895 experiments. It has been shown that artificial neural network and C5.0 decision tree models have quite similar prediction accuracy and their profitability is similar. A combination of both artificial neural network and C5.0 decision tree prediction models has been proposed. Simulations shows that the combined method is the superior one. |
| Dissertation Institution | Vilniaus universitetas. |
| Type | Master thesis |
| Language | English |
| Publication date | 2017 |