The first two chapters are dedicated to the modeling and implementation of a stock-flow consistent framework that incorporates real estate as an asset in the portfolio of the household. The third chapter investigates the main determinants of mortgage repayment of Canadian households. This first chapter presents a five-sector stock-flow consistency growth model where the portfolio decision of the households includes their choice of how much real estate they are interested in holding. The primary aim of the chapter is to model the housing market using the stock-flow consistent approach to explain the current global financial problem triggered by the housing market. The model is then simulated to predict the behaviour of various variables and propose appropriate solutions to the financial problem in the hope of returning the economy to a suitable equilibrium. Households' portfolio consists of money deposits, bills, bank equities and real estate. The other sectors that interact with the household sector are the production firms, the banks, the central bank and the government. Aside from the household sector, the banking sector ends up holding some real estate equivalent to the amount of mortgages defaulted by the households. The supply of real estate from the production sector is therefore augmented by the additional ones held by the banks. The second chapter presents the implementation of the stock-flow consistency model of first chapter. The purpose of the chapter is to run a simulation of the model and experiment with shocks to determine the path of the economic variables of the model. Another objective in performing the experiments is to find policies for mitigating the housing crisis. The model is implemented using the Eviews computer modeling software and runs until a stationary steady state is achieved. Various shocks are applied to the baseline stationary state. The results of the monetary policy show that the mortgage rate shock is more effective in influencing the growth rate of the economy as well as controlling the real estate market. Government fiscal policy is also effective in regulating the housing market. A one-period temporary fiscal policy shock is even capable of generating permanent long run growth effects. Household expectations in future housing price increases or future high rates of housing returns have the effect of heating the real estate market without comparable increases in economic growth. Policy makers must keep these expectations in check. The third chapter analyzes the determinants of mortgage repayment options in Canada. With the freedom that comes with being debt-free and owning a home one will assume that households pay off their mortgages as soon as possible. However, there are factors that inhibit households from carrying out these payoffs. The study uses Canadian micro-level data to examine factors that drive households to default, prepay or continue to make regular mortgage payments. The research methodology uses multinomial (polytomous) logistic regression analyzes. The empirical results establish that the traditional mortgage related predictor variables for repayment are statistically significant with the expected signs. The results relating to the provinces are not significantly different from each other. The results did not however provide any significance in relation to mortgage rates and the number of children in the household.

The theme of this thesis relates to solving the optimal portfolio selection problems using linear programming. There are two key contributions in this thesis. The first contribution is to generalize the well-known linear optimization framework of Conditional Value-at-Risk (CVaR)-based portfolio selection problems (see Rockafellar and Uryasev (2000, 2002)) to more general risk measure portfolio selection problems. In particular, the class of risk measure under consideration is called the Coherent Distortion Risk Measure (CDRM) and is the intersection of two well-known classes of risk measures in the literature: the Coherent Risk Measure (CRM) and the Distortion Risk Measure (DRM). In addition to CVaR, other risk measures which belong to CDRM include the Wang Transform (WT) measure, Proportional Hazard (PH) transform measure, and lookback (LB) distortion measure. Our generalization implies that the portfolio selection problems can be solved very efficiently using the linear programming approach and over a much wider class of risk measures. The second contribution of the thesis is to establish the equivalences among four formulations of CDRM optimization problems: the return maximization subject to CDRM constraint, the CDRM minimization subject to return constraint, the return-CDRM utility maximization, the CDRM-based Sharpe Ratio maximization. Equivalences among these four formulations are established in a sense that they produce the same efficient frontier when varying the parameters in their corresponding problems. We point out that the first three formulations have already been investigated in Krokhmal et al. (2002) with milder assumptions on risk measures (convex functional of portfolio weights). Here we apply their results to CDRM and establish the fourth equivalence. For every one of these formulations, the relationship between its given parameter and the implied parameters for the other three formulations is explored. Such equivalences and relationships can help verifying consistencies (or inconsistencies) for risk management with different objectives and constraints. They are also helpful for uncovering the implied information of a decision making process or of a given investment market. We conclude the thesis by conducting two case studies to illustrate the methodologies and implementations of our linear optimization approach, to verify the equivalences among four different problem formulations, and to investigate the properties of different members of CDRM. In addition, the efficiency (or inefficiency) of the so-called 1/n portfolio strategy in terms of the trade off between portfolio return and portfolio CDRM. The properties of optimal portfolios and their returns with respect to different CDRM minimization problems are compared through their numerical results.

In der Dissertation wurde ein innovatives Portfoliomodell entwickelt, welches den Präferenzen einer großen Gruppe von Investoren entspricht, die mit der traditionellen Portfolio Selektion auf Basis von Mittelwertrendite und Varianz nicht zufrieden sind. Vor allem bezieht sich die Unzufriedenheit auf eine sehr spezifische Definition der Risiko- und Wertmaße, die angenommene Nutzenfunktion, die Risikodiversifizierung sowie die Beschränkung des Assetuniversums. Dies erschwert vor allem die Optimierung der modernen Finanzprodukte. Das im Modell verwendete Risikomaß-Ausfallrisiko drückt die Präferenzen der Investoren im Bereich unterhalb der Renditebenchmark aus. Die Renditenabweichung von der Benchmark nach oben werden nicht, wie im Falle des Mittelwertrendite-Varianz-Portfoliomodells, minimiert oder als risikoneutral, wie bei dem Mittelwertrendite-Ausfallrisiko-Portfoliomodell, betrachtet. Stattdessen wird ein Wertmaß, das Chance-Potenzial (Upper Partial Moment), verwendet, mit welchem verschiedene Investorenwünsche in diesem Bereich darstellbar sind. Die Eliminierung der Annahme der normalverteilten Renditen in diesem Chance-Potenzial-Ausfallrisiko-Portfoliomodell erlaubt eine korrekte Asset Allokation auch im Falle der nicht normalverteilten Renditen, die z. B. Finanzderivate, Aktien, Renten und Immobilien zu finden sind. Bei diesen tendiert das traditionelle Mittelwertrendite-Varianz-Portfoliomodell zu suboptimalen Entscheidungen. Die praktische Anwendung des Chance-Potenzial-Ausfallrisiko-Portfoliomodells wurde am Assetuniversum von Covered Calls, Protective Puts und Aktien gezeigt. / This thesis presents an innovative portfolio model appropriate for a large group of investors which are not content with the asset allocation with the traditional, mean return-variance based portfolio model above all in term of its rather specific definition of the risk and value decision parameters, risk diversification, related utility function and its restrictions imposed on the asset universe. Its modifiable risk measure – shortfall risk – expresses variable risk preferences below the return benchmark. The upside return deviations from the benchmark are not minimized as in case of the mean return-variance portfolio model or considered risk neutral as in the mean return-shortfall risk portfolio model, but employs variable degrees of the chance potential (upper partial moments) in order to provide investors with broader range of utility choices and so reflect arbitrary preferences. The elimination of the assumption of normally distributed returns in the chance potential-shortfall risk model allows correct allocation of assets with non-normally distributed returns as e.g. financial derivatives, equities, real estates, fixed return assets, commodities where the mean-variance portfolio model tends to inferior asset allocation decisions. The computational issues of the optimization algorithm developed for the mean-variance, mean-shortfall risk and chance potential-shortfall risk portfolio selection are described to ease their practical application. Additionally, the application of the chance potential-shortfall risk model is shown on the asset universe containing stocks, covered calls and protective puts.

In der Dissertation wurde ein innovatives Portfoliomodell entwickelt, welches den Präferenzen einer großen Gruppe von Investoren entspricht, die mit der traditionellen Portfolio Selektion auf Basis von Mittelwertrendite und Varianz nicht zufrieden sind. Vor allem bezieht sich die Unzufriedenheit auf eine sehr spezifische Definition der Risiko- und Wertmaße, die angenommene Nutzenfunktion, die Risikodiversifizierung sowie die Beschränkung des Assetuniversums. Das im Modell verwendete Risikomaß-Ausfallrisiko drückt die Präferenzen der Investoren im Bereich unterhalb der Renditebenchmark aus. Die Renditenabweichung von der Benchmark nach oben werden nicht, wie im Falle des Mittelwertrendite-Varianz-Portfoliomodells, minimiert oder als risikoneutral, wie bei dem Mittelwertrendite-Ausfallrisiko-Portfoliomodell, betrachtet. Stattdessen wird ein Wertmaß, das Chance-Potenzial (Upper Partial Moment), verwendet, mit welchem verschiedene Investorenwünsche in diesem Bereich darstellbar sind. Die Eliminierung der Annahme der normalverteilten Renditen in diesem Chance-Potenzial-Ausfallrisiko-Portfoliomodell erlaubt eine korrekte Asset Allokation auch im Falle der nicht normalverteilten Renditen, die z. B. Finanzderivate, Aktien, Renten und Immobilien zu finden sind. Bei diesen tendiert das traditionelle Mittelwertrendite-Varianz-Portfoliomodell zu suboptimalen Entscheidungen. Die praktische Anwendung des Chance-Potenzial-Ausfallrisiko-Portfoliomodells wurde am Assetuniversum von Covered Calls, Protective Puts und Aktien gezeigt. / This thesis presents an innovative portfolio model appropriate for a large group of investors which are not content with the asset allocation with the traditional, mean return-variance based portfolio model above all in term of its rather specific definition of the risk and value decision parameters, risk diversification, related utility function and its restrictions imposed on the asset universe. Its modifiable risk measure – shortfall risk – expresses variable risk preferences below the return benchmark. The upside return deviations from the benchmark are not minimized as in case of the mean return-variance portfolio model or considered risk neutral as in the mean return-shortfall risk portfolio model, but employs variable degrees of the chance potential (upper partial moments) in order to provide investors with broader range of utility choices and so reflect arbitrary preferences. The elimination of the assumption of normally distributed returns in the chance potential-shortfall risk model allows correct allocation of assets with non-normally distributed returns as e.g. financial derivatives, equities, real estates, fixed return assets, commodities where the mean-variance portfolio model tends to inferior asset allocation decisions. The computational issues of the optimization algorithm developed for the mean-variance, mean-shortfall risk and chance potential-shortfall risk portfolio selection are described to ease their practical application. Additionally, the application of the chance potential-shortfall risk model is shown on the asset universe containing stocks, covered calls and protective puts.

In der Dissertation wurde ein innovatives Portfoliomodell entwickelt, welches den Präferenzen einer großen Gruppe von Investoren entspricht, die mit der traditionellen Portfolio Selektion auf Basis von Mittelwertrendite und Varianz nicht zufrieden sind. Vor allem bezieht sich die Unzufriedenheit auf eine sehr spezifische Definition der Risiko- und Wertmaße, die angenommene Nutzenfunktion, die Risikodiversifizierung sowie die Beschränkung des Assetuniversums. Dies erschwert vor allem die Optimierung der modernen Finanzprodukte. Das im Modell verwendete Risikomaß-Ausfallrisiko drückt die Präferenzen der Investoren im Bereich unterhalb der Renditebenchmark aus. Die Renditenabweichung von der Benchmark nach oben werden nicht, wie im Falle des Mittelwertrendite-Varianz-Portfoliomodells, minimiert oder als risikoneutral, wie bei dem Mittelwertrendite-Ausfallrisiko-Portfoliomodell, betrachtet. Stattdessen wird ein Wertmaß, das Chance-Potenzial (Upper Partial Moment), verwendet, mit welchem verschiedene Investorenwünsche in diesem Bereich darstellbar sind. Die Eliminierung der Annahme der normalverteilten Renditen in diesem Chance-Potenzial-Ausfallrisiko-Portfoliomodell erlaubt eine korrekte Asset Allokation auch im Falle der nicht normalverteilten Renditen, die z. B. Finanzderivate, Aktien, Renten und Immobilien zu finden sind. Bei diesen tendiert das traditionelle Mittelwertrendite-Varianz-Portfoliomodell zu suboptimalen Entscheidungen. Die praktische Anwendung des Chance-Potenzial-Ausfallrisiko-Portfoliomodells wurde am Assetuniversum von Covered Calls, Protective Puts und Aktien gezeigt. / This thesis presents an innovative portfolio model appropriate for a large group of investors which are not content with the asset allocation with the traditional, mean return-variance based portfolio model above all in term of its rather specific definition of the risk and value decision parameters, risk diversification, related utility function and its restrictions imposed on the asset universe. Its modifiable risk measure ─ shortfall risk ─ expresses variable risk preferences below the return benchmark. The upside return deviations from the benchmark are not minimized as in case of the mean return-variance portfolio model or considered risk neutral as in the mean return-shortfall risk portfolio model, but employs variable degrees of the chance potential (upper partial moments) in order to provide investors with broader range of utility choices and so reflect arbitrary preferences. The elimination of the assumption of normally distributed returns in the chance potential-shortfall risk model allows correct allocation of assets with non-normally distributed returns as e.g. financial derivatives, equities, real estates, fixed return assets, commodities where the mean-variance portfolio model tends to inferior asset allocation decisions. The computational issues of the optimization algorithm developed for the mean-variance, mean-shortfall risk and chance potential-shortfall risk portfolio selection are described to ease their practical application. Additionally, the application of the chance potential-shortfall risk model is shown on the asset universe containing stocks, covered calls and protective puts.

Disertacijoje nagrinėjama investicijų portfelio sudarymo ir valdymo rinkų dinamikos sąlygomis problematika. Globali finansų krizė parodė, kad investuojant atsiranda ne tik uždarbio galimybės, bet ir gana didelė praradimų rizika. Pagrindinis disertacijos tikslas – pasiūlyti ir empiriškai aprobuoti šiuolaikinių rinkų dinamikos iššūkius atitinkančius investicijų portfelio sudarymo ir valdymo sprendimus skirtingus investavimo polinkius turintiems investuotojams. Daktaro disertaciją sudaro įvadas, trys skyriai ir bendrosios išvados. Įvade suformuluojama mokslinė darbo problema, pagrindžiamas jos aktualumas, įvardijamas tyrimo objektas, darbo tikslas ir uždaviniai, pristatoma tyrimo metodika, darbo mokslinis naujumas ir gautų rezultatų praktinė reikšmė, įvardijami ginamieji teiginiai. Pirmajame skyriuje nagrinėjamos plačiai diversifikuoto investicijų portfelio sudarymo galimybės. Įvertinami mokslininkų pasiūlymai dėl skirtingų aktyvų (investicinio turto klasių) įtraukimo į investicijų portfelį, sudarytas biržoje prekiaujamų fondų portfelis ir įvertintas jo efektyvumas. Pasiūlytas investuotojo realiai patirtos rizikos vertinimo metodas. Antrajame skyriuje detalizuoti aktyvaus investicijų portfelio valdymo taikant finansinį svertą sprendimai. Įvertinti efektyviosios portfelių ribos pokyčiai bei aktyvaus portfelio valdymo taikant finansinį svertą tikslingumas. Pasiūlytas prognozavimo tikslumu praeityje paremtas prognozių integravimo metodas ir įvertintas jo efektyvumas integruojant... [toliau žr. visą tekstą] / The dissertation analyses the topic and problems of selection and management of investment portfolio in terms of market dynamics. The global financial crisis has revealed that investments bear not only return possibilities but also a relatively high risk of loss. The main aim of the Thesis is to propose and test empirically investment portfolio selection and management solutions matching the tendencies of modern markets for the investors with different investing preferences. The Doctoral Thesis consists of the introduction, three body chapters and conclusions. The introduction presents the scientific problem, its relevance, the object of the research, the aim and tasks of the research, methods of research, scientific novelty of the Thesis, practical significance of its results and defended statements. The first chapter provides analysis of possibilities for a widely diversified investment portfolio selection. The study of proposals of scientists on different assets combining into an investment portfolio is carried out. Portfolio of exchange traded funds is created and its efficiency is evaluated. The method for actually incurred risk evaluation is suggested. Solutions for active investment portfolio management with financial leverage are specified in the second chapter. The changes of efficient set of portfolios and expediency of active portfolio management with financial leverage are evaluated. Forecasts integration method, based on prediction accuracy in the past, is... [to full text]

The first two chapters are dedicated to the modeling and implementation of a stock-flow consistent framework that incorporates real estate as an asset in the portfolio of the household. The third chapter investigates the main determinants of mortgage repayment of Canadian households. This first chapter presents a five-sector stock-flow consistency growth model where the portfolio decision of the households includes their choice of how much real estate they are interested in holding. The primary aim of the chapter is to model the housing market using the stock-flow consistent approach to explain the current global financial problem triggered by the housing market. The model is then simulated to predict the behaviour of various variables and propose appropriate solutions to the financial problem in the hope of returning the economy to a suitable equilibrium. Households' portfolio consists of money deposits, bills, bank equities and real estate. The other sectors that interact with the household sector are the production firms, the banks, the central bank and the government. Aside from the household sector, the banking sector ends up holding some real estate equivalent to the amount of mortgages defaulted by the households. The supply of real estate from the production sector is therefore augmented by the additional ones held by the banks. The second chapter presents the implementation of the stock-flow consistency model of first chapter. The purpose of the chapter is to run a simulation of the model and experiment with shocks to determine the path of the economic variables of the model. Another objective in performing the experiments is to find policies for mitigating the housing crisis. The model is implemented using the Eviews computer modeling software and runs until a stationary steady state is achieved. Various shocks are applied to the baseline stationary state. The results of the monetary policy show that the mortgage rate shock is more effective in influencing the growth rate of the economy as well as controlling the real estate market. Government fiscal policy is also effective in regulating the housing market. A one-period temporary fiscal policy shock is even capable of generating permanent long run growth effects. Household expectations in future housing price increases or future high rates of housing returns have the effect of heating the real estate market without comparable increases in economic growth. Policy makers must keep these expectations in check. The third chapter analyzes the determinants of mortgage repayment options in Canada. With the freedom that comes with being debt-free and owning a home one will assume that households pay off their mortgages as soon as possible. However, there are factors that inhibit households from carrying out these payoffs. The study uses Canadian micro-level data to examine factors that drive households to default, prepay or continue to make regular mortgage payments. The research methodology uses multinomial (polytomous) logistic regression analyzes. The empirical results establish that the traditional mortgage related predictor variables for repayment are statistically significant with the expected signs. The results relating to the provinces are not significantly different from each other. The results did not however provide any significance in relation to mortgage rates and the number of children in the household.

The theme of this thesis relates to solving the optimal portfolio selection problems using linear programming. There are two key contributions in this thesis. The first contribution is to generalize the well-known linear optimization framework of Conditional Value-at-Risk (CVaR)-based portfolio selection problems (see Rockafellar and Uryasev (2000, 2002)) to more general risk measure portfolio selection problems. In particular, the class of risk measure under consideration is called the Coherent Distortion Risk Measure (CDRM) and is the intersection of two well-known classes of risk measures in the literature: the Coherent Risk Measure (CRM) and the Distortion Risk Measure (DRM). In addition to CVaR, other risk measures which belong to CDRM include the Wang Transform (WT) measure, Proportional Hazard (PH) transform measure, and lookback (LB) distortion measure. Our generalization implies that the portfolio selection problems can be solved very efficiently using the linear programming approach and over a much wider class of risk measures. The second contribution of the thesis is to establish the equivalences among four formulations of CDRM optimization problems: the return maximization subject to CDRM constraint, the CDRM minimization subject to return constraint, the return-CDRM utility maximization, the CDRM-based Sharpe Ratio maximization. Equivalences among these four formulations are established in a sense that they produce the same efficient frontier when varying the parameters in their corresponding problems. We point out that the first three formulations have already been investigated in Krokhmal et al. (2002) with milder assumptions on risk measures (convex functional of portfolio weights). Here we apply their results to CDRM and establish the fourth equivalence. For every one of these formulations, the relationship between its given parameter and the implied parameters for the other three formulations is explored. Such equivalences and relationships can help verifying consistencies (or inconsistencies) for risk management with different objectives and constraints. They are also helpful for uncovering the implied information of a decision making process or of a given investment market. We conclude the thesis by conducting two case studies to illustrate the methodologies and implementations of our linear optimization approach, to verify the equivalences among four different problem formulations, and to investigate the properties of different members of CDRM. In addition, the efficiency (or inefficiency) of the so-called 1/n portfolio strategy in terms of the trade off between portfolio return and portfolio CDRM. The properties of optimal portfolios and their returns with respect to different CDRM minimization problems are compared through their numerical results.

[Truncated abstract] This thesis consists of six chapters, and the main research contributions are contained in chapters two through five inclusive. The topics addressed in each chapter are distinct, but related, and the specific contributions to knowledge made by the different chapters are related to: (i) understanding more fully the nature of the demand for alcohol; (ii) explaining the relationship between reputation characteristics and consumers’ willingness to pay for wine; (iii) estimating the rate of return to Australian wine; and (iv) using financial analysis to reveal the risk diversification benefits available by including wine in an investment portfolio. The details of each contribution are briefly outlined below. Chapter 2 discusses the nature of the demand for alcohol. The demand for alcoholic beverages is an area much studied, and there are numerous studies estimating the own-price elasticity of alcoholic beverages. A review of relevant published studies indicates reported: beer own-price elasticity estimates range from -.02 to -3.00, with a mean estimate value of -.46, and standard deviation of -.41 (n = 139); wine own-price elasticity estimates range from -.05 to -3.00, with a mean estimate value of -.72, and standard deviation of .53 (n = 140); and spirits own-price elasticity estimates range from -.01 to -2.18, with a mean estimate value of -.74, and standard deviation of .47 (n = 136). Chapter 2 contributes to understanding the demand for alcohol, not by adding yet another set of elasticity estimates to an already substantial literature, but by providing a framework through which all known own-price elasticity estimates can be understood. Specifically, a meta-regression framework is employed to study previously published own-price elasticity estimates. This framework allows the effect of model design attributes to be isolated, and the underlying trend in consumer responses to price changes to be identified.

University, Diss.--Mainz, 2007.