Full metadata record
DC FieldValueLanguage
dc.contributorDepartment of Computingen_US
dc.creatorWang, Kai Kelthy-
dc.identifier.urihttps://theses.lib.polyu.edu.hk/handle/200/4972-
dc.languageEnglishen_US
dc.publisherHong Kong Polytechnic University-
dc.rightsAll rights reserveden_US
dc.titleIntelligent analysis and forecast of stock market in China : self organizing features map network with golden sectionen_US
dcterms.abstractStock is an interesting problem, there're so many people who focus on this area but not really knowing what the stock is about in China. In addition, the Chinese financial system is not matured (i.e. with defects) and the fast economic development makes the market more active than the normal market. Some capitalistic economists think of China government controlling the market unduly and do not protect the market sufficiently. I do not fully agree with this. The market has its own way most of the time unless there is some rarely large turbulence such as politics or war. We can find out some startling regularity from the history or some kind of present theories (it's mostly connecting with the Golden Section). Of course, the general index which is always presented under those quotation software can be useful if we do spend more time on research of it. But there's a same problem on it, all the methods at present are just with no guarantee, error is not a marvel thing in the market. Fortunately, we can often do prediction with effect in the longer term trend. The shorter time we do the more risk we are taking. In this paper, I try hard to pick and investigate for a valid method that can lead to better return in the security market. Due to the China market which is not mature enough, it really implies more opportunity reversible, risk and return. The propose system consist of 3 major judging standards, include Neural Network Prediction [1], the standard what based on the Golden Section Theory and Relative Strength Index. The core Network is SOM (Self-Organizing Features Map Network). It's an unsupervised learning network; this type of learning requires no teacher who knows the correct classification for the input patterns in the training set. It'll find the natural structure inherent in the input data. SOM was one of well known type of SOFM, also call it Kohonen Network. This part of prediction is based on the history term quotation which I collected before. The data is not for random, it's quite selected. It all is the stocks which have high dividend. This kind of stocks always has more active in wave fluctuation than the general stocks. This kind of data also solves the problem what the Network predictor can't accept the stocks which always play in a higher price. Another standard for the trading was Golden Section which is the second focus point in this research. It's the basis of so many famous theories, like Elliot wave theory, Gann theory etc. In the practice experiment, it really works in many testing case. Its typical point inside the trading is 0.236, 0.382, 0.5, 0.618, 0.764 etc. In different quotation, these key levels always are being the key support or pressure level. And we can evaluate the potential bottom through this theory, cooperate with the Network indicator. That's excellent set. Finally, we add a simple and effective technical index Which name Relative Strength Index assist the Network indicator. RSI [18] is an extremely popular momentum oscillator. It compares the magnitude of a stock's recent gains to the magnitude of its recent losses and turns that information into a number that ranges from 0 to 100. This index can only work with one line, really simple. And its author Wilder [27] recommends using 14 periods for the experiment.. At last, with the cooperated multi-standard system for different approaches in solving trading forecasting problems; we believe that the proposed system can benefit both the experienced and novice investors.en_US
dcterms.extentix, 112 leaves : ill. ; 30 cm.en_US
dcterms.isPartOfPolyU Electronic Thesesen_US
dcterms.issued2008en_US
dcterms.educationalLevelAll Masteren_US
dcterms.educationalLevelM.Sc.en_US
dcterms.LCSHHong Kong Polytechnic University -- Dissertations.en_US
dcterms.LCSHStock exchanges -- China.en_US
dcterms.LCSHInvestment analysis.en_US
dcterms.LCSHNeural networks (Computer science)en_US
dcterms.LCSHGolden section.en_US
dcterms.accessRightsrestricted accessen_US

Files in This Item:
File Description SizeFormat 
b22401829.pdfFor All Users (off-campus access for PolyU Staff & Students only)10.28 MBAdobe PDFView/Open


Copyright Undertaking

As a bona fide Library user, I declare that:

  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.

By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

Show simple item record

Please use this identifier to cite or link to this item: https://theses.lib.polyu.edu.hk/handle/200/4972