Dissertations / Theses on the topic 'Stocks Prices Australia'

This study deals with the estimation of the optimal hedge ratios using various econometric models. Most of the recent papers have demonstrated that the conventional ordinary least squares (OLS) method of estimating constant hedge ratios is inappropriate, other more complicated models however seem to produce no more efficient hedge ratios. Using daily AOIs and SPI futures on the Australian market, optimal hedge ratios are calculated from four different models: the OLS regression model, the bivariate vector autoaggressive model (BVAR), the error-correction model (ECM) and the multivariate diagonal Vcc GARCH Model. The performance of each hedge ratio is then compared. The hedging effectiveness is measured in terms of ex-post and ex-ante risk-return traHe-off at various forcasting horizons. It is generally found that the GARCH time varying hedge ratios provide the greatest portfolio risk reduction, particularly for longer hedging horizons, but hey so not generate the highest portfolio return.

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.
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.
Master of Information Technology by Research

This study was motivated by the desire to contribute to the understanding of the movement of house prices and the role of the so-called economic ‘fundamentals’ in the housing market, especially within an Australian context. The core objective of this thesis is to aid understanding of the economic and other mechanisms by which the Australian housing market operates. We do this by constructing an analytical framework, or model, that encompasses the most important characteristics of the housing market. This thesis examines two important aspects of the Australian housing market: movements of house prices and changes in the net capital stock of dwellings in Australia. Movements of house prices are modelled from two perspectives: firstly, using the ‘fundamental’ approach, which explains the phenomena by changes in such ‘fundamental’ explanatory variables as income, interest rates, population and prices of building materials, and secondly, by analysing spatial interdependence of house prices in Australian capital cities. Changes in stock of dwellings were also modelled on the basis of a ‘fundamental’ approach by states and for Australia as a whole...

Asset securitization is one of the most important financial innovations recently. With an impressive growth in terms of volume of issuance, from almost zero to five trillion USD, in a period of 15-20 years, it is one of the most rapidly growing markets in the financial world. Yet, little is known about this, literally invisible market. Companies engage in asset securitization for a variety of reasons and numerous advantages and disadvantages of asset securitization can be found throughout the literature. Asset securitization has an impact on a number of stakeholder groups: shareholders, managers, employees, investors, the financial markets and ultimately the overall economy and society. Asset securitization is one of the reasons for the financial crisis that started in mid 2007. Since the recent financial turmoil, it became clear the asset securitization was the primary funding source for companies in the financial industry and it was the primary supplier of credit in developed economies. Because of its importance and impact, it is very important that we study the reasons, the motivations, the consequences and the effects from this so powerful financial innovation. And it is important to study it from as many different aspects as possible. Many questions surrounding asset securitization are unanswered and it is important to answer them sooner. This study investigates the wealth effects from asset securitization on the shareholders of the securitizing companies. We study whether the announcement about a pending securitization transaction has any impact on the stock price of the securitizing company. That way we can discover whether asset securitization creates wealth, destroys wealth or has no impact on wealth at all. Not many studies have been done on this topic so far. The existing seven studies are focused mainly on the US and the EU market and report contradicting results. In this study, for the first time, data from Australia is being used. The Australian securitization market is the second, single most active securitization market in the world, after the US market. We conduct quantitative analysis on a sample of 98 securitization transactions during the period 2000-2006. With this sample, we cover almost 29% of the number of securitization transactions during that period and almost 39% in terms of volume of issuance. To analyze the data we use standard event study methodology, common for this type of studies.    Our analysis reveals that investors in Australia do not perceive asset securitization favorably. Securitizing companies’ stock price decreases in the 10 days around the securitization announcement day, resulting in statistically significant wealth losses for the originating companies’ shareholders. Furthermore, the wealth losses are significant for less frequent securitizers, for securitizers that engage in small volume securitization transactions and for securitizing companies with low asset quality.    With this study we make theoretical and practical contribution. We lend empirical support to the previous theories and we help managers, shareholders and investors shape their forecasts.

This research questions whether technical trading rules can help predict stock price movements for a sample of stocks selected from four equity markets from the Asia-Pacific region: Australia, Malaysia, Hong Kong and Thailand for the period 1989-2008. The research is split into two stages. Stage-1 of the research tests the predictability of technical trading rules against a buyand- hold strategy. The variable moving average (VMA), fixed moving average (FMA) and the trading range break (TRB) trading rules are applied to this research. Economic predictability of these rules is examined by comparing returns conditional on a trading rule buy (sell) signal against an unconditional buy-and-hold return. Any existence of excess returns can thus be established. This follows with a statistical analysis of returns using a traditional t-test methodology. Traditional statistical tests assume normally distributed returns with independent observations and a non-changing distribution across time. In Stage-2 of this research a bootstrap checks whether features such as non-normality, time-varying moments and serial correlation bias test statistics. The bootstrap involves assumptions regarding the underlying returns generating process (RGP) and allows returns conditional on a trading rule buy (sell) signal from the original stock price series to be compared with conditional returns simulated from four common null models: RW, AR (1), GARCH-M and E-GARCH models. Simulated p-values are calculated in conjunction with simulated distributions and are applied in lieu of the theoretical normal distribution. Given this process it is possible to infer as to whether non-linear dependencies in returns can be captured by any of the three trading rules. Given the null model output standard t-test outcomes of predictability of technical trading rules may be diminished and/or eliminated. Conclusions are drawn as to the predictability and profitability of the VMA, FMA and TRB trading rules when applied to the chosen stock samples. Findings of this research indicate returns conditional on technical trading rules exceed unconditional buy-and-hold returns for all stocks. Thai sample output indicates strong support in favour of the predictability of standard test results supporting the use of technical trading rules. Output for Australia, Hong Kong and Malaysia indicates that previous standard t-test outcomes of predictability may be diminished and/or eliminated. This implies that the underlying RGP may be characterised by underlying features of some/all of the stochastic models.

The research paper explores the value of information content incorporated in the first-time going concern opinion from the perspective of investors. The signaling effects of the auditors’ opinion with going concern remark issued to financially distressed companies are of a great value in case the auditor statements deliver new information content which has not been incorporated in the previously disclosed financial information. Otherwise a going concern audit opinion remains not relevant for the purpose of investors’ decision making. If the going concern audit opinion adds new information content, we gain an ability to detect a stock market reaction to the relevant public announcement. The paper examines the Australian stock market reaction to public announcements of going concern audit opinion in a short term period for the sample of the 29 first-time going concern listed companies during the 2007 to 2009 years observation period. High sample criteria are determined in order to avoid contamination effects of other price sensitive information. The impact of both the preliminary financial report and the final annual report is examined by means of the parametric and non-parametric tests aligned with the event study methodology. Consistent with previous studies in Australia, no significant financial market reaction to the final going concern audit opinion announcements inherent to the Australian environment has been found. We document that the more negative impact on the market reaction is caused by the preliminary financial report rather than the final report, which contains an audit opinion note. Correspondently, the audit opinions with going concern qualification do not add new information content for the Australian stock market participants, who base their expectations on the previously disclosed financial information.

This thesis examines the stock market price variations associated with physical asset expenditure announcements in Australia. With the exception of the study of Chen and Ho (1997) in Singapore, most capital expenditure studies in other markets investigate the announcement effects associated with changes in budgeted capital expenditures. The fact that there is almost never any firm level capital budget announcement in Australia presents a unique opportunity to examine individual physical asset expenditure announcements. ¶ Three primary hypotheses pertaining to growth opportunities, free cash flow theory, and the capital market monitoring argument are developed and tested. These arguments are formulated to explain the abnormal return variations associated with physical asset expenditure announcements. The growth opportunities hypothesis posits that the abnormal returns at physical asset expenditure announcements are positively related to a firm's growth opportunities. Both free cash flow theory and capital market monitoring hypothesis postulate that the abnormal returns at physical asset expenditure announcements are negatively related to a firm's free cash flow, and cash flow respectively. Other control explanators are incorporated from the merger and takeovers literature. ¶ Event study methodology is used to examine the abnormal returns associated with physical asset expenditure announcements. Two sets of data, intraday and daily, are used to investigate the market reaction. Intraday returns are calculated on a time-weighted approach and two methods are used to calculate intraday abnormal returns. The first method defines abnormal returns as the difference between actual returns and market returns. The second method defines abnormal returns as the difference between market-adjusted returns and market-adjusted returns on a control portfolio. Daily abnormal returns are calculated using the market model. ¶ Both univariate and multivariate analyses provide strong support for the growth opportunities hypothesis. The results suggest the quality of firms' growth opportunities is the key variable determining the direction and magnitude of the abnormal returns at announcement. Support for the capital monitoring argument and the free cash flow theory is mixed, generally with a lack of support. The free cash flow variable is found to be significantly negatively related to abnormal returns, only when a finer dummy is used in the multivariate regression. All other control variables are found to be insignificant in explaining the stock market variations once the growth opportunities variable is included in the regression. ¶ This thesis makes the following contributions. First, this thesis presents the initial empirical evidence concerning physical asset expenditure announcements in Australia. Second, the thesis shows that the quality of a firm's growth opportunities is the key factor in determining the direction and magnitude of abnormal returns around physical asset expenditure announcements. These results also suggest that the equity market in Australia reacts to physical asset expenditure announcements which contain information pertaining to growth opportunities rather than the relative size of the physical asset expenditure transactions to firm value. Third, support for the capital monitoring argument and the free cash flow theory is not strong. Fourth, all other control variables are found to be insignificant in explaining the stock market variations once market to book ratio is included in the regression. Fifth, the results suggest that prior research which fails to segregate market to book ratio and free cash flow proxy into finer partitions may have possibly underestimated the market to book and the free cash flow effects.

碩士
國立東華大學
財務金融學系
102
The purpose of this study is to examine the inter-relationship between stock and housing prices in the Australian market. To test the relationship between stock and housing prices, Granger non-causality test developed by Toda and Yamamoto (1995) is employed to investigate the causal relationship between the housing and stock prices. In addition, Engle and Granger and Johansen cointegration tests are used to analyze whether the cointegration relationship between housing and stock prices exists. In addition, this paper applies threshold autoregressive (TAR) models to explore whether the threshold effect exists. Our empirical evidence shows that no cointegration relationship and threshold effect between the Australian stock and housing price. However, housing and stock prices have a lead-lag relationship in which stock prices causes housing prices. However, the relationship is negative. The “wealth effect” is not exists. There is no significant evidence that housing prices granger cause stock prices. No significant “credit price effect” is funded in Australia market. Nevertheless, this study supports the “feedback effect” in Australia market.

Research Doctorate - Doctor of Philosophy (PhD)
The aim of this thesis is to conduct a thorough analysis of the performance of Australian banks over a long period of time, covering a period of various regulatory measures. To achieve this aim, the following four objectives are set in this thesis: first, to investigate economic efficiency (i.e. cost and profit efficiency) of the Australian banks before and after the implementation of the prudential regulation; second, to examine whether the Australian banks operate at the minimum efficient scale; third, to assess whether the efficiencies achieved contribute to wealth maximization of shareholders; fourth, to examine the determinants of Australian bank efficiency. Using a data set covering a period from 1985 through 2008, I first apply the stochastic frontier analysis (SFA) to examine the technical, cost and profit efficiency of Australian banks. A standard data envelopment analysis (DEA), as well as a slack-based DEA model (Tone 2001), is then used to assess the technical and scale efficiency of Australian banks. In addition, a Malmquist index model is used to investigate bank productivity changes over the sample period. The relationship between bank efficiency and bank stock returns is also examined using the market model. Lastly, a mixed two-step approach is used to examine efficiency and the determinants of efficiency using panel data from 1988 to 2008 across three countries, namely, Australia, Canada and the U.K.. In the first stage, a common efficiency frontier for banks in three countries is constructed including the environmental factors. The firm-level determinants of efficiency are then investigated by regressing these efficiencies on firm-specific factors. A key finding of this thesis is that, over the period from 1985 through 2008, the technical, cost and profit efficiency of Australian banks improved. However, scale efficiency showed a declining trend, which was mainly due to the scale inefficiency of the big-four banks over the sample period. Australian banks have a high level of cost and profit efficiency, but have a relatively low level of technical efficiency. Technological improvement is found to be the major driving force behind productivity changes of Australian banks, and also has a positive effect on the profit efficiency frontier. It is also observed that technical, cost and profit efficiency have a positive effect on bank stock returns, suggesting that bank efficiency is properly recognized by market participants. Compared to their regional counterparts, the big-four banks have a lower level of technical efficiency, but a higher level of cost efficiency. The low level of technical efficiency of the big-four banks is attributed to scale inefficiency. In comparison, the regional banks can achieve the same level of profit efficiency as that of the big-four banks by devising a better way of transforming inputs into outputs. Australian banks show a superior performance in terms of technical, cost and profit efficiency compared with that of Canadian and U.K. banks. The factors such as intangible assets, loans to deposits ratio and, loans to assets ratio exert a positive influence on technical efficiency. On the other hand, technical efficiency is inversely affected by size, ratio of loan loss provisions to total loans and debt to equity ratio. The findings of this thesis appear to provide justifications for the deregulatory measures and the prudential regulation framework introduced by the Australian regulatory bodies. Australian banks with increased efficiency levels and relatively high capital adequacy ratios demonstrated resilience to external shocks, such as the Asian financial crisis and the subprime mortgage crisis. An investigation of the determinants of bank efficiency suggests that an Australian bank manager has the choice of tuning up either the capital structure or the asset structure to improve efficiency. However, these findings should be interpreted with caution due to the limitations relating to data unavailability and efficiency evaluation techniques.

Portfolio theory, created by economists, was a breakthrough in financial economics. This theory looks at the stock market as a whole and analyses how, for a given rate of expected return, assets can be invested efficiently and how risk can be minimized. An effectively diversified portfolio minimizes the unsystematic risk which is affected by factors that are specific to the individual firms and, to some extent, the industry in which the firm operates. The unsystematic risk is, therefore, manageable by diversification. The systematic risk, however, cannot be managed by a simple approach of diversification. Despite the fact that there are many other factors contributing to the systematic risk of a portfolio, the risk and return of a diversified portfolio is mainly affected by domestic and overseas economic factors.

A general result from theoretical and empirical research in financial market is that information, market microstructure and trading clientele affect prices and trading patterns. Previous research, however, concentrated mainly on larger well traded security markets. This thesis extend this research to the thinly traded and informationally dependent Australian All Ordinaries Index (AOI), the Share Price Index (SPI) future contract and the arbitrage pricing series between these two markets. Time series and transfer function techniques are applied to intraday and interday data to examine the impact of information flow and trading structures on trading and price patterns and the spillover effect across markets. The empirical evidence from this the is thesis is consistent with a number of complex and dynamic relationship which vary across trading times and market place. Reults indicate that in individual AOI and SPI markets, structural trading halts, price setting mechanism, and the arrival of information at market opening are associated with price overreaction and higher volatility, but this is not the case in the arbitrage price series. At the close of trading there are significant average price deviations in the individual and arbitrage erie, but without any excess price volatility. Unexpected intraday trading activity in the futures market preceded price exchange in the stock and future markets; the arrival of information has a greater impact on future price and the short term volatility in the futures markets is significantly higher than in the stock market. These influence in the future market, however, did not lead to any spillover effect which increased the long term volatility of the stock market. However, there is evidence of sustained and predictable mispricing in the SPI index futures arbitrage series with mean reversion in the arbitrage series being a function of different time of the day and possible market psychology. On the other hand, significant mean reversion is associated with increased trading volume and efficient transaction cost bound argument. Overall, the research in this thesis indicate that market are affected by a mixture of information, trading micro structure and subtle market reaction, which are both rational and irrational. The major conclusion is that price and volume reaction are complex and flexible theories are required to explain the intricate working of the marketplace. These are important consideration to be borne in mind by policy makers, regulator and market traders.