published_or_final_version / abstract / Statistics and Actuarial Science / Master / Master of Philosophy

Man Po Kong = 信貸相聯票據和住宅按揭的預期損失之估值 / 文普綱. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2004. / Includes bibliographical references (leaves 85-86). / Text in English; abstracts in English and Chinese. / Man Po Kong = Xin dai xiang lian piao ju he zhu zhai an jie de yu qi sun shi zhi gu zhi / Wen Pugang. / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- The Structural model --- p.3 / Chapter 2.1 --- Merton's model --- p.3 / Chapter 2.2 --- The term structure of interest rate --- p.7 / Chapter 2.3 --- The default-triggering mechanism and derivations from strict priority rule --- p.9 / Chapter 2.4 --- Stationary leverage ratio --- p.11 / Chapter 2.5 --- The three-factor structural model --- p.12 / Chapter 3 --- Credit-linked Notes with early default risk --- p.18 / Chapter 3.1 --- Introduction to credit-linked notes --- p.18 / Chapter 3.2 --- The pricing of credit-linked notes --- p.20 / Chapter 3.3 --- Non mean-reverting leverage ratios --- p.21 / Chapter 3.3.1 --- Special case (pQv=0) --- p.23 / Chapter 3.4 --- Mean reverting leverage ratios --- p.25 / Chapter 4 --- Numerical results and discussion --- p.28 / Chapter 4.1 --- Exact solution (KQ=kv=PQv=PVr=0) --- p.31 / Chapter 4.2 --- "Lower bound approximation (kQ,kv≠0,pQr,pvr≠0)" --- p.37 / Chapter 4.2.1 --- Effect of interest rate --- p.43 / Chapter 4.3 --- Monte Carlo simulation (PQV≠0) --- p.47 / Chapter 5 --- Expected loss of residential mortgage loans --- p.56 / Chapter 5.1 --- Introduction to residential mortgage loans --- p.56 / Chapter 5.2 --- Calculation of expected loss of residential mortgage loans --- p.59 / Chapter 6 --- Numerical results and discussion --- p.65 / Chapter 6.1 --- Numerical results --- p.65 / Chapter 7 --- Conclusion --- p.73 / Chapter A --- Methodology --- p.75 / Chapter A.1 --- Monte Carlo Simulation --- p.76 / Chapter A.2 --- Finding lower and upper bound approach --- p.79 / Chapter A.2.1 --- Single stage approximation --- p.79 / Chapter A.2.2 --- Multistage lower bound approximation --- p.82 / Bibliography --- p.85

by Leung Tsun Ip. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2006. / Includes bibliographical references (leaves 72-78). / Abstracts in English and Chinese. / Chapter 1. --- Introduction --- p.1 / Chapter 2. --- Literature Review --- p.6 / Chapter 2.1 --- Development of estimation of Value-at-Risk (VaR) --- p.6 / Chapter 2.2 --- Methods to evaluate VaR --- p.9 / Chapter 2.2.1 --- Non-paremetric Method --- p.9 / Chapter 2.2.2 --- Semi-parametric Method --- p.11 / Chapter 2.2.3 --- Parametric Method --- p.12 / Chapter 3. --- Extreme Value Theory (EVT) --- p.16 / Chapter 3.1 --- Introduction of Extreme Value Theory (EVT) --- p.16 / Chapter 3.1.1 --- Block Maxima Approach --- p.18 / Chapter 3.1.2 --- Peaks over Threshold (POT) Approach --- p.21 / Chapter 3.1.3 --- Comparison between Block Maxima and POT Approach --- p.22 / Chapter 3.2 --- Numerical Illustration --- p.23 / Chapter 3.2.1 --- Data --- p.23 / Chapter 3.2.2 --- Diagnosis --- p.24 / Chapter 4. --- Dynamic Extreme Value Theory (DEVT) --- p.29 / Chapter 4.1 --- Theoretical Framework of DEVT --- p.29 / Chapter 4.2 --- Estimation of Parameters --- p.32 / Chapter 4.3 --- Determination of Threshold Level --- p.37 / Chapter 4.4 --- Estimation of zq --- p.44 / Chapter 5. --- Backtesting and Time Aggregation --- p.49 / Chapter 5.1 --- Backtesting DEVT --- p.49 / Chapter 5.2 --- Time Aggregation --- p.55 / Chapter 6. --- Case Study: China Aviation Oil Singapore (CAO) Incident --- p.61 / Chapter 6.1 --- Background Information --- p.61 / Chapter 6.2 --- Data Analysis --- p.63 / Chapter 6.3 --- Suggestion --- p.68 / Chapter 7. --- Discussion --- p.71 / References --- p.72 / Chapter A. --- Appendix --- p.79

Yeung Kin Bong. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (leaves 69-73). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Credit risk modeling --- p.3 / Chapter 1.2 --- Uniqueness of bi-directional: hybrid system --- p.4 / Chapter 1.3 --- Scope of the study --- p.5 / Chapter 2 --- Literature Review --- p.6 / Chapter 2.1 --- Statistical / Empirical approach --- p.6 / Chapter 2.2 --- Structural approach --- p.8 / Chapter 3 --- Background --- p.10 / Chapter 3.1 --- Merton structural default model --- p.10 / Chapter 3.2 --- Cross-sectional regression analysis (CRA) --- p.15 / Chapter 3.3 --- Neural network learning (NN) --- p.16 / Chapter 3.3.1 --- Single-layer network --- p.17 / Chapter 3.3.2 --- Multi-layer perceptron (MLP) --- p.20 / Chapter 3.3.3 --- Back-propagation network --- p.22 / Chapter 3.3.4 --- "Supervised, unsupervised and combine unsupervised-supervised learning" --- p.23 / Chapter 3.4 --- Weaknesses of uni-directional modeling --- p.23 / Chapter 4 --- Methodology --- p.26 / Chapter 4.1 --- Bi-directional modeling --- p.26 / Chapter 4.2 --- Asset price estimation --- p.31 / Chapter 4.3 --- Quantifying accounting data noise --- p.33 / Chapter 5 --- Proposed Model --- p.37 / Chapter 5.1 --- Core of the model --- p.37 / Chapter 5.2 --- Feature selection --- p.41 / Chapter 5.3 --- Bi-directional default neural system --- p.44 / Chapter 6 --- Implementations --- p.49 / Chapter 6.1 --- Data preparation --- p.50 / Chapter 6.2 --- Experiment --- p.51 / Chapter 6.3 --- Empirical results --- p.61 / Chapter 6.3.1 --- Predicted spreads from the uni-directional models --- p.61 / Chapter 6.3.2 --- Predicted spreads from the proposed bi-directional model --- p.63 / Chapter 6.3.3 --- Performance comparison --- p.64 / Chapter 7 --- Conclusions --- p.67 / Bibliography --- p.69

The classical mean-variance model treats the upside and downside equally as risks. This feature is undesirable, in the eyes of a profit-making investor. In this regard, the downside Lower Partial Moments (LPM) are more attractive as alternative risk measures, since they only penalize the downside. This thesis is mainly concerned with the issues related to downside risk measures. We consider two different environments, under which our investigations shall proceed. The first one is the world of Q-radial distributions. The Q-radial distributions generalize the normal distribution and uniform distribution, among many other useful classes of probability distributions. The second type of setting that we will investigate assumes that the distribution of the assets' returns is ambiguous, and the only available (and reliable) knowledge that we have is the first few moments of the distribution. In the first setting, we show that if the investment return rates follow a Q-radial distribution, then the LPM related Risk Adjusted Performance Measures (RAPM), such as the Sortino ratio, the Omega Statistic, the upside potential ratio, and the normalized LPM, are all equivalent to the ordinary Sharpe ratio, which is easy to compute and optimize. Conversely, if all normalized LPM's are equivalent to the Sharpe ratio, then the underlying distribution must be Q-radial. Therefore, this property characterizes the class of Q-radial distributions in which the Sharpe ratio is essentially the only risk adjusted performance measure. If the distribution is unspecified, and only the first few moments (first, second, and/or fourth) are known, we develop tight upper bounds on the lower partial moment E[(r -- X+m], where r ∈ reals and X is stochastic. Based on such tight bounds we then consider the corresponding robust portfolio selection problem, in which the distribution of the investment return is ambiguous, but its first few moments are assumed to be known. We show that if the first two moments are known and the risk measures are either the lower partial moments or the Conditional Value-at-Risk (CVaR), then the optimal portfolio is mean-variance efficient. Moreover, one can formulate the (adjustable) two-stage robust portfolio selection problem as a convex program with finite representations. If more than two moments are known, then the problem is NP-hard in general. In that case we consider approximative models instead. We then proceed to consider the problem of how to alleviate regrets in a decision problem when the parameters are ambiguous, or part of the information will only become known in a dynamic fashion. Since the models we consider in this thesis are mostly in the minimax format, we also consider a general minimax model and study a progressive finite representation approach, which can be used to prove the minimax theorem constructively without any fixed-point theorem or hyperplane separation theorems. / Chen, Li. / Adviser: Shuzhong Zhang. / Source: Dissertation Abstracts International, Volume: 70-06, Section: B, page: 3762. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (leaves 106-111). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest dissertations and theses, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

Jin Hanqing. / "July 2004." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2004. / Includes bibliographical references (p. 133-139). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest dissertations and theses, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

Wong, Chi Ying. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (leaves 71-74). / Abstracts in English and Chinese. / Abstract Page --- p.ii / Acknowledgment Page --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Literature Review --- p.5 / Chapter 3 --- Review of minimax portfolio selection model --- p.11 / Chapter 3.1 --- The I∞ model --- p.11 / Chapter 4 --- Portfolio optimization with group investment limits --- p.16 / Chapter 4.1 --- The model --- p.16 / Chapter 4.2 --- The optimal investment strategy --- p.17 / Chapter 4.2.1 --- All assets are risky --- p.18 / Chapter 4.2.2 --- Some riskfree assets are involved --- p.39 / Chapter 4.3 --- Chapter summary --- p.40 / Chapter 5 --- Tracing out the efficient frontier --- p.41 / Chapter 5.1 --- Properties of the efficient frontier --- p.42 / Chapter 5.2 --- The algorithm --- p.51 / Chapter 5.3 --- Time complexity of the algorithm --- p.56 / Chapter 5.4 --- Chapter summary --- p.57 / Chapter 6 --- Finding the investor's optimal portfolio --- p.58 / Chapter 6.1 --- Investor's portfolio with given A --- p.58 / Chapter 6.2 --- Chapter summary --- p.60 / Chapter 7 --- Numerical experiments --- p.61 / Chapter 7.1 --- Finding the efficient frontier numerically --- p.61 / Chapter 7.2 --- Performance between mean-variance model and I∞ model --- p.64 / Chapter 7.2.1 --- Data analysis --- p.64 / Chapter 7.2.2 --- Experiment description and discussion --- p.65 / Chapter 7.3 --- Chapter summary --- p.67 / Chapter 8 --- Conclusion --- p.68 / Bibliography --- p.71 / Appendix --- p.75 / Chapter A --- Stocks for finding the efficient frontiers with and without bound constraints --- p.75 / Chapter B --- List of companies --- p.77 / Chapter C --- Graphical Results --- p.81

published_or_final_version / Statistics and Actuarial Science / Doctoral / Doctor of Philosophy

The profitability of a manufacturer could be largely affected by underlying uncertainties embedded in the fast-changing business environment. Random factors, such as input material price at the procurement end or output product price and demand at the sales end, might produce significant risks. Effective financial hedging therefore needs to be taken to mitigate these risk exposures. Although it is common to use commodity futures to control the risks at either end separately, little has been done on the hedging of these risk exposures in an integrated manner. Therefore, this study aims to develop a planning approach that performs financial hedging on both the procurement risk and the sales risk in a joint manner. This planning approach is based on a framework that has a risk-averse commodity processor that procures input commodity and sells output commodity in the spot market, while hedging the procurement risk and sales risk through trading futures contracts in the commodity markets. Both the input and output commodities futures are used for the hedging. A both-end-hedging model is developed to quantitatively evaluate the approach. The evaluation is based on an objective function that considers both profit maximisation and risk mitigation. Decisions on spot procurement, input futures hedging position, and output futures hedging position are optimised simultaneously. As the input commodity is the main production material for the output commodity, positive correlation between the input material price and the output product price is considered. The customer demand is considered negatively correlated with the output product price. An ethanol plant using corn as the main input material is employed as an example to implement the proposed model. The model is represented as a stochastic program, and the Gibson-Schwartz two-factor model is employed to describe the stochastic commodity prices. Historical commodity price data are used to estimate the parameters for the two-factor model with state-space form and Kalman filter. By generating various scenarios representing evolving prices and the random customer demand, the stochastic program could be solved using linear programming algorithms under its deterministic equivalent. Numerical experiments are carried out to demonstrate the benefit that could be gained from applying the both-end-hedging approach proposed in this study. Comparing with traditional no-hedging model or single-end-hedging models, the improvement obtained from the proposed model is found to be significant. The effectiveness of the model is further tested in various price trend and price correlation, demand elasticity and volatility, and risk attitude of the decision maker. It is found that the proposed approach is robust in these various circumstances, and the approach is especially effective when the price trend is uncertain and when the decision maker has a strong risk-averse attitude. / published_or_final_version / Industrial and Manufacturing Systems Engineering / Master / Master of Philosophy

published_or_final_version / Statistics and Actuarial Science / Doctoral / Doctor of Philosophy