This research investigated the feasibility and capability of neural network-based approaches for predicting the direction of the Australian Stock market index (the target market). It includes several aspects: univariate feature selection from the historical time series of the target market, inter-market analysis for finding the most relevant influential markets, investigations of the effect of time cycles on the target market and the discovery of the optimal neural network architectures. Previous research on US stock markets and other international markets have shown that the neural network approach is one of most powerful techniques for predicting stock market behaviour. Neural networks are capable of capturing the non-linear stochastic and chaotic patterns in the stock market time series data. This study discovered that the relative return series of the Open, High, Low and Close prices of the target market, show 6-day cycles during the studied period of about 14 years. Multi-layer feedforward neural networks trained with a backpropagation algorithm were used for the experiments. Two major testing methods: testing with randomly selected test data and forward testing, were examined and compared. The best neural network developed in this study has achieved 87%, 81% 83% and 81% accuracy respectively in predicting the next-day direction of the relative return of the Open, High, Low and Close prices of the target market. The architecture of this network consists of 33 input features, one hidden layer with 3 neurons and 4 output neurons. The best input features set includes the relative returns from 1 to 6 days in the past of the Open, High, Low and Close prices of the target market, the day of the week, and the previous day’s relative return of the Close prices of the US S&P 500 Index, US Dow Jones Industrial Average Index, US Gold/Silver Index, and the US Oil Index. / Master of Information Technology by Research

This research investigated the feasibility and capability of neural network-based approaches for predicting the direction of the Australian Stock market index (the target market). It includes several aspects: univariate feature selection from the historical time series of the target market, inter-market analysis for finding the most relevant influential markets, investigations of the effect of time cycles on the target market and the discovery of the optimal neural network architectures. Previous research on US stock markets and other international markets have shown that the neural network approach is one of most powerful techniques for predicting stock market behaviour. Neural networks are capable of capturing the non-linear stochastic and chaotic patterns in the stock market time series data. This study discovered that the relative return series of the Open, High, Low and Close prices of the target market, show 6-day cycles during the studied period of about 14 years. Multi-layer feedforward neural networks trained with a backpropagation algorithm were used for the experiments. Two major testing methods: testing with randomly selected test data and forward testing, were examined and compared. The best neural network developed in this study has achieved 87%, 81% 83% and 81% accuracy respectively in predicting the next-day direction of the relative return of the Open, High, Low and Close prices of the target market. The architecture of this network consists of 33 input features, one hidden layer with 3 neurons and 4 output neurons. The best input features set includes the relative returns from 1 to 6 days in the past of the Open, High, Low and Close prices of the target market, the day of the week, and the previous day’s relative return of the Close prices of the US S&P 500 Index, US Dow Jones Industrial Average Index, US Gold/Silver Index, and the US Oil Index. / Doctor of Philosophy

This research investigated the feasibility and capability of neural network-based approaches for predicting the direction of the Australian Stock market index (the target market). It includes several aspects: univariate feature selection from the historical time series of the target market, inter-market analysis for finding the most relevant influential markets, investigations of the effect of time cycles on the target market and the discovery of the optimal neural network architectures. Previous research on US stock markets and other international markets have shown that the neural network approach is one of most powerful techniques for predicting stock market behaviour. Neural networks are capable of capturing the non-linear stochastic and chaotic patterns in the stock market time series data. This study discovered that the relative return series of the Open, High, Low and Close prices of the target market, show 6-day cycles during the studied period of about 14 years. Multi-layer feedforward neural networks trained with a backpropagation algorithm were used for the experiments. Two major testing methods: testing with randomly selected test data and forward testing, were examined and compared. The best neural network developed in this study has achieved 87%, 81% 83% and 81% accuracy respectively in predicting the next-day direction of the relative return of the Open, High, Low and Close prices of the target market. The architecture of this network consists of 33 input features, one hidden layer with 3 neurons and 4 output neurons. The best input features set includes the relative returns from 1 to 6 days in the past of the Open, High, Low and Close prices of the target market, the day of the week, and the previous day’s relative return of the Close prices of the US S&P 500 Index, US Dow Jones Industrial Average Index, US Gold/Silver Index, and the US Oil Index. / Master of Information Technology by Research

This research investigated the feasibility and capability of neural network-based approaches for predicting the direction of the Australian Stock market index (the target market). It includes several aspects: univariate feature selection from the historical time series of the target market, inter-market analysis for finding the most relevant influential markets, investigations of the effect of time cycles on the target market and the discovery of the optimal neural network architectures. Previous research on US stock markets and other international markets have shown that the neural network approach is one of most powerful techniques for predicting stock market behaviour. Neural networks are capable of capturing the non-linear stochastic and chaotic patterns in the stock market time series data. This study discovered that the relative return series of the Open, High, Low and Close prices of the target market, show 6-day cycles during the studied period of about 14 years. Multi-layer feedforward neural networks trained with a backpropagation algorithm were used for the experiments. Two major testing methods: testing with randomly selected test data and forward testing, were examined and compared. The best neural network developed in this study has achieved 87%, 81% 83% and 81% accuracy respectively in predicting the next-day direction of the relative return of the Open, High, Low and Close prices of the target market. The architecture of this network consists of 33 input features, one hidden layer with 3 neurons and 4 output neurons. The best input features set includes the relative returns from 1 to 6 days in the past of the Open, High, Low and Close prices of the target market, the day of the week, and the previous day’s relative return of the Close prices of the US S&P 500 Index, US Dow Jones Industrial Average Index, US Gold/Silver Index, and the US Oil Index. / Doctor of Philosophy

Thesis (Ph. D.)--University of North Texas, Dec., 2007. / Title from title page display. Includes bibliographical references.

No description available.

In this work, we investigate different techniques to predict the monthly trend direction of the S&P 500 market index. The techniques use a machine learning classifier with technical and macroeconomic indicators as input features. The Support Vector Machine (SVM) classifier was explored in-depth in order to optimize the performance using four different kernels; Linear, Radial Basis Function (RBF), Polynomial, and Quadratic. A result found was the performance of the classifier can be optimized by reducing the number of macroeconomic features needed by 30% using Sequential Feature Selection. Further performance enhancement was achieved by optimizing the RBF kernel and SVM parameters through gridsearch. This resulted in final classification accuracy rates of 62% using technical features alone with gridsearch and 60.4% using macroeconomic features alone using Rankfeatures

by Ng Wai-Leung Anthony. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1999. / Includes bibliographical references (leaves 83-87). / Abstracts in English and Chinese. / Chapter I. --- INTRODUCTION --- p.1 / Chapter II. --- LITERATURE REVIEW --- p.5 / Chapter 2.1 --- The Importance of Stock Index and Index Futures --- p.6 / Chapter 2.2 --- Importance of Index Forecasting --- p.6 / Chapter 2.3 --- Reasons for the Lead-Lag Relationship between Stock and Futures Markets --- p.9 / Chapter 2.4 --- Importance of the lead-lag relationship --- p.10 / Chapter 2.5 --- Some Empirical Findings of the Lead-Lag Relationship --- p.10 / Chapter 2.6 --- New Approach to Financial Forecasting - Artificial Neural Network --- p.12 / Chapter 2.7 --- Artificial Neural Network Architecture --- p.14 / Chapter 2.8 --- Evidence on the Employment of ANN in Financial Analysis --- p.20 / Chapter 2.9 --- Hong Kong Securities and Futures Markets --- p.25 / Chapter III. --- GENERAL GUIDELINE IN DESIGNING AN ARTIFICIAL NEURAL NETWORK FORECASTING MODEL --- p.28 / Chapter 3.1 --- Procedure for using Artificial Neural Network --- p.29 / Chapter IV. --- METHODOLOGY --- p.37 / Chapter 4.1 --- ADF Test for Unit Root --- p.38 / Chapter 4.2 --- "Error Correction Model, Error Correction Model with Short- term Dynamics, and ANN Models for Comparisons" --- p.38 / Chapter 4.3 --- Comparison Criteria of Different Models --- p.39 / Chapter 4.4 --- Data Analysis --- p.39 / Chapter 4.5 --- Data Manipulations --- p.41 / Chapter V. --- RESULTS --- p.42 / Chapter 5.1 --- The Resulting Models --- p.42 / Chapter 5.2 --- The Prediction Power among the Models --- p.45 / Chapter 5.3 --- ANN Model of Input Variable Selection Using Contribution Factor --- p.46 / Chapter VI. --- CAUSALITY ANALYSIS --- p.54 / Chapter 6.1 --- Granger Casuality Analysis --- p.55 / Chapter 6.2 --- Results Interpretation --- p.56 / Chapter VII --- CONSISTENCE VALIDATION --- p.61 / Chapter VIII --- ARTIFICIAL NEURAL NETWORK TRADING SYSTEM --- p.67 / Chapter 7.1 --- Trading System Architecture --- p.68 / Chapter 7.2 --- Simulation Runs using the Trading System --- p.77 / Chapter XI. --- CONCLUSIONS AND FUTURE WORKS --- p.79

After a long-time preparation, CSI 300 index futures has made a milestone in the financial market in China in the 16 of April, 2010. In order to know what kind of impact will bring to stock market after the appearance of stock index future, the study discusses volatility and volume separately. On one hand, the study applies Modified Levene and GJR-GARCH as the empirical model, and the result indicates that stock return fluctuation is a short-term phenomenon. However, the result shows that the stock return volatility has no difference in the long-run. Furthermore, it not only reduces the asymmetric return fluctuation from good and bad news cause but improve the information efficiency in the spot market after the introduction of the stock index futures. On the other hand, the study applies multiple regression model and panel model to examine the crowding-out effect and the volume difference after the stock index futures enters the market. First, there is no crowding-out effect in the stock market. Second, both the trading volume of the constituent and non-constituent stocks increase after the introduction of the stock index futures, whereas the level of increasing trading volume of the constituent stocks is larger than non- constituent stocks are.

This thesis aims to provide detailed investigation into the role and functioning of the FTSE-100 stock index futures contract, by examining four interrelated issues. Chapter 1 reviews the literature, demonstrating that stock index futures can increase investor utility by offering hedging and investment opportunities. Further, the price discovery role of futures is discussed. Chapter 2 investigates the risk return relationship for the FTSE-100 contract within a CAPM framework. While CAPM adequately explains returns prior to October 1987, post-crash the contract is riskier and excess returns and a day of the week effect are evident. Chapter 3 examines the impact of futures on the underlying spot market using GARCH, which allows examination of the link between information and volatility. While spot prices are more volatile post-futures, this is due to more rapid impounding of information. The view that futures destabilise spot markets and should be subject to further regulation is questioned. Chapter 4 examines futures market efficiency using the Johansen cointegration procedure and variance bounds tests which are developed here. Results suggest futures prices provide unbiased predictions of future spot prices for 1, 2 and 4 months prior to maturity of the contract. For 3, 5 and 6 months prior to maturity the unbiasedness hypothesis does not hold. Chapter 5 discusses the major role of futures; hedging. Hedge ratios and hedging effectiveness are examined in relation to duration and expiration effects. Hedge ratio stability is also examined. Finally, hedging strategies based on historical information are examined. Results show there are duration and expiration effect, hedge ratios are stationary and using historical information does not greatly reduce hedging effectiveness. The FTSE-100 contract is shown to be a highly effective means by which to hedge risk. Chapter 6 provides a summary and concluding remarks concerning the relevance of the research carried out here.