無 / We derive the MacLaurin series for the moments of the idle time with respect to the parameters in the service time and interarrival time distributions for a GI/G/I queue. The light traffic derivatives are obtained to investigate the quality of a well-known MacLaurin series. The coefficients in these series are expressed in terms of the derivatives of the interarrival time density function evaluated at zero and the moments of the service time distribution, which can be easily calculated through a simple recursive procedure. The result for the idle period is easily taken as input to the calculation of other performance measures of the system, e.g., interdeparture time distributions. Numerical examples are given to illustrate these results.

This article investigates the detection and identification problems for changing of regimes about non-linear time series process. We apply the concept of genetic algorithm and AIC criterion to test the changing of regimes. This way is different from traditional detection methods According to our statistical decision procedure, the mean of moving average and the genetic detection for the underlying time series will be considered to decide change points. Finally, an empirical application about the detection and identification of change points for the Taiwan Business Cycle is illustrated.

The central problem addressed by this thesis is to attempt and reconcile our experience of time with our scientific understanding of time. Science tells us that time is static yet we experience it as dynamic. In the literature there tend to be two positions. Those who follow the science and claim that time is static and that our experience is mind-independent; those who favour our experience and question the science. I attempt to reconcile these positions. To do this I adopt terminology set out by McTaggart (1908) who termed the static view the B series and the dynamic view the A series. The literature that has developed out of this breaks down into the A Theory where time is the past, present and future; and the B Theory, where time is just involves events being earlier than or later than other events. I reject both positions as accounts of ontology. I adopt McTaggart’s C series, a series of betweenness only, on the grounds that it is this series that is mostly aligned to science. Given the C series, our experience requires explanation. A claim of mind-dependency is insufficient. I argue that the A series really refers to mind-dependent features that are brought out by our interaction with the C series; much like the way that colour is brought out by our interaction with a colourless world. The B series is the best description of the contents of time, not time itself. To examine the experience of time I adopt phenomenology to describe that experience. From within experience I show that certain features of that experience cannot be attributed to a mind-independent reality and use this as further evidence for the above claims. Finally I suggest that most theories of time are driven by the view that a theory of time has to be consistent. I examine recent developments in logic to see whether such a consistent requirement is needed. I conclude that the most we can get out of paraconsistent approaches is inconsistent experiences, not inconsistent reality. I conclude that the A series is the best description of our experience of time, the C series the best description of the ontology of time, and the B series as the best description of the contents of time. This reconciles our experience with our understanding of time. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1286776 / Thesis(PhD)-- School of Humanities, 2007

The huge volume of time series data generated in many applications poses new challenges in the techniques of data storage, transmission, and computation. Further more, when the time series are in the form of streaming data, new problems emerge and new techniques are required because of the streaming characteristics, e.g. high volume, high speed and continuous flowing. Approximate representation is one of the most efficient and effective solutions to address the large-volume-high-speed problem. In this thesis, we propose a dynamic representation scheme for streaming time series. Existing methods use a unitary function form for the entire approximation task. In contrast, our method adopts a set of function candidates such as linear function, polynomial function(degree ≥ 2), and exponential function. We provide a novel segmenting strategy to generate subsequences and dynamically choose candidate functions to approximate the subsequences. / Since we are dealing with streaming time series, the segmenting points and the corresponding approximate functions are incrementally produced. For a certain function form, we use a buffer window to find the local farthest possible segmenting point under a user specified error tolerance threshold. To achieve this goal, we define a feasible space for the coefficients of the function and show that we can indirectly find the local best segmenting point by the calculation in the coefficient space. Given the error tolerance threshold, the candidate function representing more information by unit parameter is chosen as the approximate function. Therefore, our representation scheme is more flexible and compact. We provide two dynamic algorithms, PLQS and PLQES, which involve two and three candidate functions, respectively. We also present the general strategy of function selection when more candidate functions are considered. In the experimental test, we examine the effectiveness of our algorithms with synthetic and real time series data sets. We compare our method with the piecewise linear approximation method and the experimental results demonstrate the evident superiority of our dynamic approach under the same error tolerance threshold.

Thesis (M.Ed.)--University of Hong Kong, 1984. / Includes bibliographical references (leaf 106-112). Also available in print.

Thesis (Ph. D.)--University of Hong Kong, 2007. / Title proper from title frame. Also available in printed format.

Thesis (Ph. D.)--University of Hong Kong, 2007. / Title proper from title frame. Also available in printed format.

<p>This thesis consists of three self-contained essays in the fields of banking and portfolio choice.</p><p>Banking and Optimal Reserves in an Equilibrium Model:</p><p>I address the question of reserves in banking, particularly the fact that reserves are substantially larger than the stipulated reserve requirements by Bank of International Settlements. My contribution is to show that when the underlying values of borrowers are correlated, banks should hold positive reserves, regardless of the regulation. I use a derived distribution for debt portfolios to show that intermediation in a debt market will outperform direct lending, even if intermediaries are allowed to default. The model used is a generalization of Williamson (1986), with Costly State Verification as asymmetric information. By using a factor model for the value of entrepreneurs' projects, I introduce a positive probability for banks to default. It is shown that, in equilibrium, banks choose to hold capital reserves that are almost large enough to eliminate the expected auditing cost for their depositors. The reason is that auditing does not provide any utility and hence, the cake to be split between banks and depositors is enlarged by reserves as an insurance against bad outcomes. It is also shown that the more correlation there is in the debt portfolio, the larger is the optimal reserve level. This could explain why small regional banks in Sweden often have more than twice the reserve level of their nation-wide competitors.</p><p>Optimal Rebalancing of Portfolio Weights under Time-varying Return Volatility:</p><p>This paper considers horizon effects on portfolio weights under time varying and forecastable return volatility. The return volatility is modeled as a GARCH-M, which is sufficiently general to encompass both constant and time varying means. The analysis confirms earlier results, namely that there are no horizon effects when the stochastic process, which governs asset returns, has a constant mean. However, when time varying and forecastable volatility is included in the mean equation, there are horizon effects. I show three features to be of importance for the horizon effect: First, the size of the parameter on conditional volatility in the mean equation and second, persistence in conditional volatility. Third, the asymmetry in volatility has some effect. In addition, the parameter of relative risk aversion is important. For low levels of risk aversion, only very small effects on portfolio weights are present; when the level of risk aversion increases, so does the effects on portfolio weights. Portfolio weights increase for the first 2-3 years when the investment horizon is increased; the total effect slightly exceeds 10%.</p><p>Can Parameter Uncertainty Help Solve the Home Bias Puzzle?</p><p>A well-known puzzle in international finance is the equity home bias. This paper illustrates a mechanism where exchange rate estimation risk causes equity home bias. Estimation risk is introduced into a standard mean-variance portfolio framework by having return time series with different lengths. We argue that the exchange rate return history, which is a part of the local currency return on a foreign investment, is likely to be substantially shorter than the available return histories of equity indices due to, for example, exchange rate regime shifts. To econometrically deal with return histories of different lengths we utilize a framework devised by Stambaugh (1997). The impact of estimation risk on an optimal portfolio is tested with data from Sweden and the U.S. Our results suggest that explicitly accounting for estimation risk causes the domestic investor to increase his fraction of domestic assets. While the introduction of exchange rate estimation risk is not powerful enough to explain the whole home bias observed in data, the results of this paper illustrate a mechanism that is often overlooked in discussions of international portfolio diversification.</p>

<p>Economic growth around the world has not been equal for a long time. Some economics grow faster while others grow slower. But economists have predicted that the slower growing economics will eventually converge with the faster growing economy as some point in the future. This is known as the convergence hypothesis. In this study, we test this hypothesis for Ghana and the Western Europeans countries with UK been a proxy for these countries, using time series data to determine whether or not it holds. We determine how fast or slow this convergence process is by using the returns to scale concept on Ghana’s economy and latter account for factor that determines economic growth in sectors. The study supported the null hypothesis of convergence i.e. Ghana is catching up with the Western European countries. The study also shown that Ghana growth accounting exhibit decreasing returns meaning convergence is relatively slow and also signifies that Ghana is not on a balanced growth path (this refers to the simultaneous, coordinated expansion of several sectors of the economy). The study showed a negative relationship between GDP and labour both in the long run and short run relationship. Again the study showed a positive relationship between GDP and capital, Agric and Industrial sector. Lastly, the study showed a negative relationship between GDP and AID and Service in the long run and positive relationship in the short run.</p>

In spite of their importance in many systems, liquid surfaces have been explored at the microscopic level to a much lesser extent than solids. Most surface analysis must take place in vacuum, a major drawback for liquids. The technique of time-of-flight scattering and recoil spectrometry (TOF-SARS) has been applied to molecular liquid surfaces for the first time. The apparatus borrows key elements from previous TOF-SARS experiments on solids and from molecular beam scattering (MBS) and features excellent surface specificity and the ability to detect all elements. A high-vacuum time-of-flight spectrometer was developed for the purpose of measuring the surface atomic concentration of atoms in low-vapor pressure liquid samples, and hence to infer preferred surface orientations. The TOF-SARS experiment involves surface bombardment with inert gas ions in the 1-3 keV energy range. During the interaction surface atoms may either (a) induce scattering of primary ions or (b) recoil from the surface. A binary collision model describes the kinematics and dynamics of the interactions well, allowing prediction of velocities and probabilities of particles leaving the surface. Particles that reach a detector along a ~1.1 m flight path are separated by velocity, and signals are collected as a histogram, revealing relative measured intensities that are converted to ratios of accessible surface atoms. Comparing the measured atomic ratios with computer-simulated accessible atomic ratios for various possible orientations gives insight into preferred surface orientation. A number of systems were explored m this work: liquids including a complementary pair of molecules having distinct 'head-tails' structures; glycerol as a highly H-bonded system, and a room-temperature molten salt. Preliminary results reveal that surface molecules appear in most cases to adopt some preferred orientation at the interface. The TOF-SARS technique was able to distinguish 'head' from 'tail' in molecules exhibiting that structure, suggesting only part of the head was accessible. In glycerol, all but two possible orientations were ruled out but the symmetrical nature of the molecule prohibits definitive assignment. The ionic liquid was found to have the cation and anion sharing the surface population roughly equally, and a preferred orientation for the substituted aromatic anion was discovered. / Graduation date: 2000