In analysing the extent to which adequate research work may have been undertaken in the specific area of computer disaster prevention, it was found that little work had been done. In the real-life situation, it was also concluded that, in the vast majority of cases, no adequate disaster prevention controls were in use at IT installations. Guidance for the analysis and management of the risk associated with computer disasters, as a result, has also been inadequate and lacking in uniformity, specially in the areas of risk identification and risk entities interactions and relationships. This research has involved developing and delivering a methodology which would help IT risk managers in implementing effective computer disaster prevention controls. A knowledge based system (KBS) approach has been used to build a prototype system which provides full support in this important area of decision making, and to show how the representation of risks can be handled.

This thesis addresses the need to reduce inefficiencies in management of insurance company risk capital. The laxity in managing the cost of capital is a result of dysfunctional property/casualty risk classification and capital accumulation practices in the insurance industry. We reclassify risk based on both peril and financial functional features, in order to capture all the facets of risk affecting a firm and ultimately to achieve optimal capital allocation. With the purpose of reducing inefficiencies in mind, we explore and isolate the impact of regulation on insurance company profitability. We use barrier option pricing models to mimic the impact of solvency requirements on firm-wide risk. This methodology of measuring risk is better than plain vanilla option pricing models, in that, through the option to an early default, we are able to capture the economic significance of financial distress, and allocate firm-wide risk capital. The firm-wide risk is incidentally used to empirically test the impact of risk on the cost of carry, the quality of operational profitability and forward asset commitment per unit of liabilities. Our empirical test confirms a strong relationship between firm-level risk, and the cost of carry, return on policyholders' surplus and the cost of capital per contract underwritten. The results are better than previous results obtained using plain vanilla option-pricing models and reveal the importance of incorporating solvency requirements in defining the economic significance of insolvency. The results also points to the importance of advised risk classification procedures to the whole process of integrated risk measurement and financing, which we explore in this study.

A number of risks have been proposed in the literature to be associated with hyperfunctional voice disorder (HVD), one of the most common communication disorders. Yet, it is not distributed randomly in the population, certain population groups are at higher risks of developing voice disorders. It is generally agreed that the development of voice disorders involves a multifactorial genesis. The study of risks has been documented in different diseases and also in the World Health Report (W.H.O., 2002). The probabilistic approach has been recommended to effectively manage the likelihood of health outcome in relations to disease development (Tonetti, 1988) and systematically devise prevention and intervention programs targeting population at risk. Yet, in the study of the development of HVD, the lack of a universally agreed theoretical framework prohibited the establishment of such structure and research on advancement on preventive programs. The present thesis aimed to investigate the adoption of the FMAT risk assessment framework based on the probabilistic approach (WHO) to the field of hyperfunctional voice disorders. A Voice Risk Calculator (VRC) Questionnaire was developed focusing on the vocal loading, physiological/medical and psycho-emotional indicators and all subjects completed this questionnaire. The VRC Questionnaire was then validated based on the FMAT framework using a cross-sectional study was used to identify risk indicators associated with HVD development in the local population and a longitudinal study was employed to validate these risk indicators as risk factors. A total of 192 Cantonese-speaking subjects participated in the cross-sectional study including 123 dysphonic subjects and 69 non-dysphonic control subjects and 7 in the longitudinal study. Instrumental measurements including the voice range profile, aerodynamic measurements and the Voice Activity and Participation Profile (VAPP, Ma & Yiu, 2001) were also used as part of the validation procedure. The findings showed that significant differences were found between the dysphonic and non-dysphonic group in the cross-sectional study based on results from the instrumental measurements protocol. A minimal set of selected VRC questionnaire items were also determined (Items 1, 3, 25 and somatization scale) to differences between the subject groups in this study. Thus a set of locally-applicable risk indicators have been suggested. Yet, only minimal changes have been detected in a high-risk group targeted in the longitudinal study. Research (Beck, 1994) indicated that disease progression takes over a time frame of at least more than two years. Thus the small subject size and temporal element of the longitudinal study in the present thesis limited research aim to be achieved. Nonetheless that first phase of the FMAT framework for hyperfunctional voice disorders have been established in the current study and a finalized version of the Voice Risk Calculated Questionnaire has been developed for future research. / published_or_final_version / Speech and Hearing Sciences / Doctoral / Doctor of Philosophy

Regarding the relationships among different insurance claims, especially in non-life insurance, the dependence behaviour in various models has been studied extensively. In this thesis, some discrete-time risk models with dependence structures would be investigated. One traditional discrete-time risk model is the time series risk model, in which the dependence would be on two aspects: time correlated claims and dependent business classes. A general vector (multivariate) autoregressive moving average (VARMA) model would be adopted to analyze the ruin probability of a surplus process. An upper bound for the ruin probability is derived for the general order of multivariate time series models in claims. Simulation studies are carried out for model comparison for finite time ruin probabilities. Another class of risk model is the compound binomial risk model, where the dependence structure would be based on the existence of a so-called by-claim in the claim process. The by-claim could be incurred in the same period as the main insurance claim, or it would be incurred in the next period, depending on a certain probability. A randomized dividend payment scheme with some fixed threshold value in surplus level would also be considered in this thesis. A methodology is discovered to obtain the Gerber-Shiu expected penalty function for the extended model. The final model investigated in this thesis is the periodic time series risk model. The periodic structure of the model gives a practical interpretation of the business cycle, in which there are high season and low season for the business. Some lower order periodic time series models are considered for the claim structures. / published_or_final_version / Statistics and Actuarial Science / Doctoral / Doctor of Philosophy

Background: The epidemic of childhood obesity is of increasing public health concern, with major implications for long-term health. Prevention strategies are urgently needed. Most of the evidence concerning risk factors for childhood obesity comes from observational studies, mainly from Western populations. In the West, socio-economic position (SEP) is often associated with potential risk factors and with childhood obesity, making these observational studies open to residual confounding. Evidence from a setting with a different confounding structure can be valuable in disentangling whether associations observed in Western settings reflect potentially reversible causal effects of risk factor or are confounded by SEP. Objectives: This thesis took advantage of a large (n=8327), population-representative Chinese birth cohort from a developed non-Western setting, Hong Kong, where the confounding structure between potential risk factors and childhood obesity is different, to examine the association of four modifiable risk factors (mode of delivery, the timing of solid food introduction, type of child care and dairy product consumption) with adiposity from infancy to early puberty. Methods: Adiposity from infancy to early puberty was proxied by age- and sex-specific body mass index (BMI) standardized scores (z-scores) from 3 months to 13 years, relative to the 2006 World Health Organization (WHO) child growth standards for 0-5 years and the 2007 WHO growth reference for 5-19 years. Overweight (including obesity) was defined according to International Obesity Task Force cut-off. I compared three marginal models (maximum likelihood estimation, generalized estimating equations and quantile regression) to ascertain the optimal way of modeling the population-averaged association of early life risk factors with BMI z-score because of the complex data structure with inevitably some missing data. All three methods were used to examine the adjusted associations of mode of delivery and the timing of solid food introduction with BMI z-score from infancy to early puberty and with overweight (including obesity) from early childhood to early puberty. Multivariable linear and logistic regression were used to examine the adjusted associations of the type of child care at 6 months, 3 years, 5 years and 11 years with BMI z-score and overweight (including obesity) at 13 years, and the association of dairy product consumption at 11 years with BMI z-score at 13 years. Results: My analyses were robust to the choice of marginal model. Mode of delivery, the timing of solid food introduction and dairy product consumption were not associated BMI z-score or overweight (including obesity), but informal child care was associated with higher BMI z-score and overweight at early puberty. Conclusions: In this population-representative birth cohort from an understudied non- Western developed setting with little patterning of childhood adiposity by SEP, informal child care (by family members and/or in-home employed help) may be a target for intervention. Conversely, cesarean section, early introduction of solid food and lack of dairy product consumption do not appear to be contributing to the current obesity epidemic. Non-replication in a different context suggests some observed associations in the West may be indicators of residual confounding rather than of causality. / published_or_final_version / Public Health / Doctoral / Doctor of Philosophy

In the literature of ruin theory, there have been extensive studies trying to generalize the classical insurance risk model. In this thesis, we look into two particular risk processes considering multi-dimensional risk and dependent structures respectively. The first one is a bivariate risk process with a dividend barrier, which concerns a two-dimensional risk model under a barrier strategy. Copula is used to represent the dependence between two business lines when a common shock strikes. By defining the time of ruin to be the first time that either of the two lines has its surplus level below zero, we derive a discrete approximation procedure to calculate the expected discounted dividends until ruin under such a model. A thorough discussion of application in proportional reinsurance with numerical examples is provided as well as an examination of the joint optimal dividend barrier for the bivariate process. The second risk process is a semi-Markovian dual risk process. Assuming that the dependence among innovations and waiting times is driven by a Markov chain, we analyze a quantity resembling the Gerber-Shiu expected discounted penalty function that incorporates random variables defined before and after the time of ruin, such as the minimum surplus level before ruin and the time of the first gain after ruin. General properties of the function are studied, and some exact results are derived upon distributional assumptions on either the inter-arrival times or the gain amounts. Applications in a perpetual insurance and the last inter-arrival time before ruin are given along with some numerical examples. / published_or_final_version / Statistics and Actuarial Science / Master / Master of Philosophy

Traditionally, in the context of ruin theory, most judgements are made on an immediate sense. An example would be the determination of ruin, in which a business is declared broke right away when it attains a negative surplus. Another example would be the decision on dividend payment, in which a business pays dividends whenever the surplus level overshoots certain threshold. Such scheme of decision making is generally being criticized as unrealistic from a practical point of view. The Parisian concept is therefore invoked to handle this issue. This idea is deemed more realistic since it allows certain delay in the execution of decisions. In this thesis, such Parisian concept is utilized on two different aspects. The first one is to incorporate this concept on defining ruin, leading to the introduction of Parisian ruin time. Under such a setting, a business is considered ruined only when the surplus level stays negative continuously for a prescribed length of time. The case for a fixed delay is considered. Both the renewal risk model and the dual renewal risk model are studied. Under a mild distributional assumption that either the inter arrival time or the claim size is exponentially distributed (while keeping the other arbitrary), the Laplace transform to the Parisian ruin time is derived. Numerical example is performed to confirm the reasonableness of the results. The methodology in obtaining the Laplace transform to the Parisian ruin time is also demonstrated to be useful in deriving the joint distribution to the number of negative surplus causing or without causing Parisian ruin. The second contribution is to incorporate this concept on the decision for dividend payment. Specifically, a business only pays lump-sum dividends when the surplus level stays above certain threshold continuously for a prescribed length of time. The case for a fixed and an Erlang(n) delay are considered. The dual compound Poisson risk model is studied. Laplace transform to the ordinary ruin time is derived. Numerical examples are performed to illustrate the results. / published_or_final_version / Statistics and Actuarial Science / Master / Master of Philosophy

Not available / text

Not available / text

Not available / text