Please use this identifier to cite or link to this item: https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/155
Title: Use Stock Market Data to assist investors on Getting more accurate transaction decision
Authors: Chandima, A.V.L.
Issue Date: 14-Oct-2013
Abstract: Colombo Stock Exchange (CSE) is one of the best performing stock exchanges in South Asia. Presently lots of investors (i.e. people who buy or sell stocks) connected with the CSE. Generally stock market domain is a dynamic and unpredictable environment. However stock trading (buy / sell) process doesn’t exist without prediction. In the current stock trading process investors predict the following day stock price solely based on the stock broker’s assistant. Stock brokers are the one who directly connects with the stock trading process regularly. Therefore they have massive experience and ongoing knowledge, so they can use this to assist the investors. But current stock trading process in CSE, stock brokers do not have much time to completely disclose their knowledge and assist investors uniformly. Also in some situations stock brokers are not aware of several patterns based on the previous data, but they are more valuable for the investors to do their stock trading process. So most investors are unable to win their stock trading process and gain profit. As the solution for the above mentioned problems above titled system was introduced. It can assist the investors in their stock trading process more accurately than the past. Mainly there are three methods used for stock market prediction depend on several procedures. Among those one of the most widely used method is technical analysis, which uses technical indicators and correlated charts based on the past stock data to predict market movements and to identify trading opportunities. Technical indicators are dividing into two parts called leading indicators (gives a trading signal before the new trend and reversal occurs) and lagging indicators (gives a trading signal after the trend has started). Past researches confirms that using them separately was not successful in stock prediction. So proposed system was found combination of lagging and leading technical indicators allows substantially increase profitability of a trading system rather using them separately. So this thesis proposes a supervised learning approach to generate trading signals using a combination of technical indicators. Proposed system was combined one leading indicator called Relatively Strength Index and two lagging indicators called Bollinger Bands and Simple Moving Average. Technical indicators calculation and related chart generating were automatically done by the proposed system developed tool add-ins for Microsoft Excel. Finally investors can analyze this charts according to the rules defined in the system. This analytical knowledge is very valuable for Investors to get accurate transaction decision with minimum broker assist.
URI: http://hdl.handle.net/123456789/155
Appears in Collections:Master of Computer Science - 2012

Files in This Item:
File Description SizeFormat 
Final Desertation.pdf
  Restricted Access
2.77 MBAdobe PDFView/Open Request a copy


Items in UCSC Digital Library are protected by copyright, with all rights reserved, unless otherwise indicated.