This preview shows page 1. Sign up to view the full content.
Unformatted text preview: s or small
Recurrent neural network approach
Kamijo and Tanigawa  propose the use of an Elman recurrent net for predicting the
future stock prices using extracted features from past daily high, low and closing stock
prices. The method used tries to extract triangle patterns in stock prices which are seen
graphically by plotting the daily high, low and closing prices. A triangle refers to the
beginning of a sudden stock price rise after which the high and low prices appear and the
price oscillates for some period before the lines converge. The neural network is trained to
recognize such triangle patterns in the stock prices. As such, it is mainly a categorization
approach to recognize whether a pattern is a triangle or not. Such knowledge can be
useful in judging whether a price rise is permanent.
Prediction based on past stock values and other fundamental indicators
The second category of research assumes that other fundamental factors such as present
interest rates, bond rate and foreign exchange rates affect...
View Full Document
This note was uploaded on 07/20/2012 for the course ECON 203 taught by Professor Girishdev during the Spring '12 term at Indian Institute of Technology, Kharagpur.
- Spring '12