Dissertation > Excellent graduate degree dissertation topics show

The effectiveness of technical analysis based on neural network research

Author: TangYuHong
Tutor: ZengYong
School: University of Electronic Science and Technology
Course: Quantitative Economics
Keywords: Neural Network Momentum Strategies Reverse strategy Excess returns The projected performance
CLC: F830.91
Type: Master's thesis
Year: 2005
Downloads: 302
Quote: 0
Read: Download Dissertation

Abstract


Technical analysis, financial analysis as an important part of the means with the development of information technology is constantly inject new theoretical support for more great, new technologies, new applications of the theory of technical analysis has been an unprecedented development. Since the establishment of the Chinese stock market, based on technical analysis studies have increasingly attracted the attention of investors. In this paper, based on the technical analysis of the neural network prediction model in China's stock market efficiency. Firstly, the technical analysis research and development status quo, systematic introduction to the core content of the effectiveness of financial markets, technical analysis of the validity of the concept of assumptions. By the citation of the existing research has answered the significance of the technical analysis of the validity of the securities market in China. On the basis of the above theory, this paper presents the analysis and forecasting model based on artificial neural network technology, the use of artificial neural network feedforward network infrastructure, the use of back-propagation algorithm to study and train to fit the predicted securities yields. Been testing the amount of on technical analysis gains significant nature of the test is not subject to the prior setting assuming the conditions of the problem and the presence of data mining problems (DATA-snooping), this paper adopted Bootstrap method to to test the statistics of the conclusions significant resistance and robust nature. Empirical studies using single hidden layer BP neural network as a predictive model and the random walk model as compared the research momentum strategy applied to the Shanghai A-share Composite Index excess returns. The results showed that: the random walk model can not obtain excess returns, the momentum strategy based on neural network to obtain the total excess return is positive; use Bootstrap inspection method to get the p-value that momentum strategy based on neural network has good profitability and stability. In addition, by calculating and comparing also found higher prediction accuracy further shows the characteristics of technical analysis tools based on neural network is superior to simple technical analysis methods. An Empirical Study II uses the same hidden layer BP neural network to predict the yield on the reverse strategy, and the sample divided into three represents a different market environment sub-interval examine the different effects of the reverse strategy. The results showed that: the reverse strategy based on neural network in the whole sample period to get a positive excess return, the use of the Bootstrap test methods p-value also proved that these excess returns significantly greater than zero, it can be inferred applied to the reverse strategy based on neural network China's securities market is forecast to have a certain degree of profitability. Further analysis of the different market environment: reverse strategy can play a very good effect in the consolidation of the city state the City rising effect has been reduced, the city continued to fall largely ineffective, it also shows that investors in the process technical analysis needs to take into account different market conditions. Overall conclusions of this paper support analysis based on neural network technology in China's stock market at this stage, can predict stock returns.

Related Dissertations

  1. Development of the Platform for Compressor Optimization Design and Aerodynamic Optimization Design in the Transonic Compressor,TH45
  2. Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
  3. Study on Virtual Detector of Infrared Hyper-Spectral Image,TP391.41
  4. The Application of Fuzzy Control and Neural Network in Planar Double Inverted Pendulum,TP273.4
  5. Research of Visual Servoing in 4-DOF Robotic Manipulator,TP242.6
  6. Research on Visual Servo System of Mechanical ARM,TP242.6
  7. Design of Weapon Detection Device Control System,TP183
  8. Municipal tourism land use planning environmental impact assessment,X820.3
  9. Research on the Inteligent System of High Performance Concrete Mix Design in Zhujiang Triangle Area,TU528
  10. Study on Taste Characteristic of Taste Peptide Enzymatic Production from Oyster Base on A Neural Network Method,TS254.4
  11. Smart Control Research in Paddy Drying Based on BP Neural Network,S226.6
  12. BP network optimization based on genetic algorithm optimization of the biodiesel process,TE667
  13. Fire Fighting System Research for the Offshore Platform,U698.4
  14. The Research on Evaluation of Living Status Systems of Expressway Relocated People,D523
  15. Research on Inspection Technology of Dehydrated Garlic Slice Based on Computer Vision,TP391.41
  16. Research and Application of Safety Assessment Method Based on Neural Network,X937
  17. The Design of the Coal Mine’s Safety Evaluation System Based on Wavelet Neural Network,TD79
  18. Study of Gray Neural Network Model of Enterprise’s Safety Devotion,X913.4
  19. Research of VRLA Battery On-line Montoring and Contorl System,TP277
  20. The Identification of Fault Type in Transmission Lines Based on Neural Network,TP183
  21. Enterprise Security Benefit Evaluation and Development Strategies,F272;F224.5

CLC: > Economic > Fiscal, monetary > Finance, banking > Finance, banking theory > Financial market > Securities market
© 2012 www.DissertationTopic.Net  Mobile