Forecasting Stock Price Using Neuro-Fuzzy
This study aims to study the stock price forecasting using Neuro-Fuzzy. This is a collaboration between Neural Networks and Fuzzy Logic. The data were from the Stock Exchange of Thailand, the period from 1 January 2009 until December 31, 2011, The data be divided into two parts, the first part is used for learning and the second part will be put to the testing. The data used for learning is a input of 50 data, 100 data, 200 data, 300 data, 400 data and 500 data before forecasting. The test data used for forecasting, it is a one month, 3 months, 5 months and 9 months in the second series will begin on 1 March 2011. The results showed that a input of 500 data are error less than others. The study concluded that the error is reduced when the input data set has a lot more. It may not always, because the direction of movement of information, learning and testing data sets that moving in opposite directions. The forecast was not accurate enough for use excessive levels of profitability. The data in past can not to predict future stock prices. Therefore the Stock Exchange of Thailand is an indication that the efficient market.
- There are currently no refbacks.
Copyright (c) 2016 วารสารวิทยาการจัดการ (Journal of Management Sciences)
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
วารสารวิทยาการจัดการ (Journal of Management Sciences) ISSN: 0125-8362