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

Univ., Diss.--Amsterdam.

Thesis (doctoral)--Université de Lausanne.

Thesis (Ph. D.)--Ohio State University, 2003. / Title from first page of PDF file. Document formatted into pages; contains x, 104 p.; also includes graphics (some col.). Includes bibliographical references. Available online via OhioLINK's ETD Center.

Thesis (M.B.A.)--University of Hong Kong, 1991.

Thesis (Ph. D.)--Georgia Institute of Technology, 2004. / UMI number: 3199305. Includes bibliographical references (leaves 278-289). Also available online (World Wide Web)

Thesis (doctoral)--Universiẗat Frankfurt (Main), 2005. / Includes bibliographical references (p. 421-464).

[pt] Visando equacionar o abastecimento do mercado asfalto brasileiro e atendimento das metas financeiras da empresa, evitando custos excessivos com estoques, a presente dissertação propõe um modelo de gestão de estoque de asfaltos para a companhia analisada. O modelo desenvolvido busca melhorar a atual gestão que é feita com base na experiência dos colaboradores, trazendo cálculos e fundamentos acadêmicos para as práticas de uma unidade de produção. O trabalho tem como principais entregas: cálculo de estoques de segurança trimestrais, propostos devido a sazonalidade da demanda (apresentando os potenciais ganhos financeiros em relação ao anual) e levando em consideração as incertezas da demanda e lead time de produção, sendo o nível de serviço limitado em função da tancagem; definição da política de controle de estoques, contendo o cálculo do ponto de ressuprimento; e um comparativo entre os custos de estoque atual e teórico calculado. Além disso, como produto final, foi elaborada e entregue uma planilha de cálculos para uso exclusivo da companhia analisada para que o estoque de segurança e o ponto de ressuprimento possam ser atualizados à medida que as variáveis de entrada sofram alterações. / [en] Aiming to balance market supply and financial goals, avoiding excessive costs with inventories, this dissertation proposes a model of stock management of asphalt for the enterprise in study. The developed management model aims to enhance the current management model, which is based on employees and collaborators experience, introducing measurements and academic grounds to the enterprise practices. This assessment aims to deliver the calculation of the safety stock levels on a quarterly basis instead of an annual basis due to the seasonality of the demand (the potential gains of quarterly calculation compared to the annul calculation will also be presented) and considering the uncertainties of production demand and lead time, the level of service being limited due to the capacity to stock, the definition of the stock control policies, the calculation of the replenishment point, and a comparison between current and theoretical stock costs, including as final product, an automatic spreadsheet for the exclusive use of the analyzed company, to update the variables and calculations when necessary.

This diploma thesis deals with stock {--} exchange business on European continent, especially on organised markets. Main goal is to underline czech stock market, its financial derivatives. In the theoretical part chosen general topics concerning stock-exchange business on organized markets are characterized. Importace of stock {--} exchange, it structure, principles of business, history of stock {--} exchange business are described. Next chapter deals with czech market (organised Prague stock exchange and an organiser of off-exchange trading RM {--} system), its comparison, the new financial derivatives market as well. As compared markets are chosen: London, Wien, Warsaw, Budapest, Italian stock Exchange, Deutsche Borse, cross border exchange organisation NYSE EURONEXT and Prague stock exchange. The application part is based upon information gathered from web sites of particular stock exchanges and international organisations. Chosen figures concerning positron of stock exchange markets on international market with securities are: organizational structure, way of trading and clearing, trading hours, distributing information, trading financial derivatives, statistical indicators: number of members, market capitalizaton, trade volume, number of transactions and evaluation of futures trades.

This study investigates FTSE/JSE All Share index monthly and daily equity returns for evidence of the January and TY effect. Four different measures of monthly return are analysed for the 1995-2006 period, whilst daily returns are analysed during the 1995-2005 period. In addition to this, analysis is conducted on monthly Fama-MacBeth risk premium estimates tor the FTSE/JSE All Share Index. Descriptive statistics are first analysed, followed by ANOV A or Kruskai-Wallis tests, the paired t-test and finally dummy variable regression analysis in investigating the seasonality of FTSE/JSE All Share Index returns and risk premia. Analysis on monthly returns reveals an absence of the January effect, however a positive slightly statistically significant December effect is found. Thus, investors earn abnormal returns on equity during the month of December. The results from the Fama-MacBeth risk premia estimates reveals highly statistically significant negative risk premia seasonal patterns during March, July and September. Thus, investors are in fact penalised for investing in equities during these months. In addition, the analysis reveals an absence of a December effect in risk premia, which contradicts the risk-return trade-off central to modem finance. The daily return analysis reveals a highly significant Turn-of-the-Year effect (TY), which suggests that investors earn abnormal returns on days at the turn of the year. Therefore, it is concluded that a December effect is apparent in South African equity monthly returns, whilst a March, July and September effect is apparent in South African equity risk premia contradicting the risk-return trade-off central to modem finance. In addition to this, a TY effect is present in South African equity daily returns.