This project has three parts. The first part is to use the efficient frontier and find the tangency portfolio to form our optimal portfolio. We built our portfolio using the Interactive Brokers software and rebalanced every week for 4 holding periods to see the relationship between our projected returns and actual market returns. In the second part we considered the Capital Asset Pricing Model (CAPM) and ran linear regressions on the stocks we chose in the first part of the project. This process is based on our idea of finding the systematic risk in each stock to improve our stock choosing ability. In the last part we introduce the concept of factor models and add more factors into our original CAPM model. Via a back-testing method, we test the reasonability of our factors and give advice to further improve our portfolio optimization project.

Long term asset creates more revenue, however it is riskier in a liquidity sense. Our question is: How does a liquidity constrained bank make decisions between profitability and liquidity? We present a computable DSGE model of banks optimal portfolio choices under liquidity constraints. Our theory predicts that liquidation plays an important role in a bank's portfolio model. Even though liquidation is an off-equilibrium phenomenon, banks can have rich loan portfolios due to the possibility of liquidation. Liquidity condition is a key factor in banks portfolio. In a moderate liquidity situation, a bank can lend more profitable longer term loans, however, if a shock in liquidity is expected, then the bank lends more loans in short term. According to the liquidity conditions, the bank can have medium term loans which are different from other previous literature. In addition, we extend our model to the bank's securities business where the bank's debts are largely short term deposit. Our theory predicts that the bank securities business produces a chasm between a real liquidity of economy and market liquidity. Banks can have more liquidity by selling their securitized loans, and as our model already pointed out, a good liquidity condition makes the bank have more profitable but less liquid long term loans. As a consequence, long term loans are accumulated with this securitization, simply because a long term loan gives higher revenue. Any market turbulence can invoke a problem in economy wide liquidity.

Dans la premiere partie de cette these, nous donnons un point de vue financier sur l'etude des taux d'interet long terme. En finance, les modeles classiques de taux ne s'appliquent plus pour des maturites longues (15 ans et plus). Nous montrons que les techniques de maximisation d'utilite esperee permettent de retrouver la regle de Ramsey (qui relie la courbe des taux a l'utilite marginale de la consommation optimale). En marche incomplet, il est possible de montrer un analogue de la regle de Ramsey et nous examinons la maniere dont la courbe des taux est modifiee. Ensuite nous considerons le cas ou il y a une incertitude sur un parametre du modele, puis nous etendons ces resultats au cas ou les fonctions d'utilites sont stochastiques. D'autre part nous proposons dans cette these une nouvelle maniere d'apprehender la consommation, comme des provisions que l'investisseur met de cote pour les utiliser en cas d'un evenement de defaut. Alors le probleme de maximisationn de l'utilite esperee de la richesse et de la consommation peut etre vu comme un probleme de maximisation de l'utilite esperee de la richesse terminale avec un horizon aleatoire. La deuxieme partie de cette these concerne l'analyse stochastique des processus ponctuels determinantaux. Les processus determinantaux et permanentaux sont des processus ponctuels dont les fonctions de correlations sont donnees par un determinant ou un permanent. Les points de ces processus ont respectivement un comportement de repulsion ou d'attraction: ils sont tres loin de la situation d'absence de correlation rencontree pour les processus de Poisson. Nous etablissons un resultat de quasi-invariance: nous montrons que si nous perturbons les point le long d'un champ de vecteurs, le processus qui en resulte est toujours un determinantal, dont la loi est absolument continue par rapport a la distribution d'origine. En se basant sur cette formule et en suivant l'approche de Bismut du calcul de Malliavin, nous donnons ensuite une formule d'integration par parties.

In this thesis we study models, which search for an optimal portfolio from a set of stocks. On the contrary to the classical approach focusing only on expected return and variance, we examine models where an additional crite- rion of skewness is included. Furthermore we formulate a model for measuring performance of a portfolio defined as the distance from the Pareto efficient frontier. In numerical experiments we apply the models on historical prices and stock data from the electronic stock market NASDAQ. We analyze the stock data from companies listed in the index NASDAQ-100. We conclude by comparing of optimal portfolios created using different models among each other, with trivial single-stock portfolios and the with NASDAQ-100 index itself.

This master thesis is focused on portfolio diversification. In the Czech Republic, the majority of the population still deposits their free funds to current accounts, but the yield is not sufficient to cover the devaluation caused by inflation. In addition, investments in securities enable these funds to be better valued (naturally with a higher risk). The aggregate of all investments is called the investment portfolio. Harry Markowitz is the founder of modern portfolio theory. The aim of the thesis is to compile an optimal portfolio from chosen financial assets. The theoretical part of the thesis describes the terms such as the financial market, its nature and function and the basic elements of the investment strategy - profitability, risk and liquidity. On top of that, this part describes problems of portfolio theory with a focus on the Markowitz model of optimization. In total 15 stocks-issuing companies are selected from various industries. These companies are traded both on the Czech and American stock markets. The practical part is focused on creating optimal portfolio of selected financial assets. For different attitudes of the investor to risk and its selected strategy the optimal portfolio according to Markowitz is compiled. The weights of individual securities are determined as well as the yield and risk of the portfolios created and an effective boundary is demarcated.

This thesis contributes to the econometric literature in two ways. Firstly, it introduces a new multivariate count model that presents advances in several aspects. Our multivariate time series count model can deal with issues of discreteness, overdispersion (variance greater than the mean) and both cross- and serial correlation, all at the same time. We follow a fully parametric approach and specify a marginal distribution for the counts where, conditionally on past observations the means follow a vector autoregressive process (VAR). This enables to attain improved inference on coefficients of exogenous regressors relative to the static Poisson regression, while modelling the serial correlation in a flexible way. The method is also innovative in the use of copulas, which builds the dependence structure between variables with given marginal distributions. This makes it possible to model the contemporaneous correlation between individual series in a very flexible way. Secondly, this thesis introduces a new approach to estimate the multivariate reduced rank regressions when the normality assumption is not satisfied. We propose to use the copula tool to generate multivariate distributions and, we show that this method can be applied in multivariate settings. In terms of financial literature, this thesis provides two contributions. Firstly, with our multivariate count model we analyze diverse market microstructure issues about the submission of different types of orders by traders on stock markets. With this model, we can fully take into account the interactions between submissions of the various types of orders, which represent an advantage with respect to univariate models such as the autoregressive conditional duration model. Secondly, it contributes to portfolio research proposing a new dynamic optimal portfolio allocation model in a Value-at-Risk setup. This model allows for time varying skewness and kurtosis of portfolio distributions and the model parameters are estimated by weighted maximum likelihood in an increasing window setup. This last property allows us to have more accurate portfolio recommendations in terms of the amount to invest in the risk-free interest rate and in the risky portfolio.

This paper considers dynamic optimal portfolio strategies of utility maximizing investors in the presence of risk constraints. In particular, we investigate the optimization problem with an additional constraint modeling bounded shortfall risk measured by Value at Risk or Expected Loss. Using the Black-Scholes model of a complete financial market and applying martingale methods we give analytic expressions for the optimal terminal wealth and the optimal portfolio strategies and present some numerical results.

The paper investigates dynamic optimal portfolio strategies of utility maximizing portfolio managers in the presence of risk constraints. Especially we consider the risk, that the terminal wealth of the portfolio falls short of a certain benchmark level which is proportional to the stock price. This risk is measured by the Expected Utility Loss. We generalize the findings our previous papers to this case. Using the Black-Scholes model of a complete financial market and applying martingale methods, analytic expressions for the optimal terminal wealth and the optimal portfolio strategies are given. Numerical examples illustrate the analytic results.

Thesis (MSc (Mathematics))--Stellenbosch University, 2008. / In the Benchmark Approach to Finance, it has been shown that by taking the Growth Optimal Portfolio as numéraire, a candidate for a pricing derivatives formula under the real world probability can be given. This result allows us to price in an incomplete financial market model. The result comes from two different approaches. In the first approach we use the supermartingale property of portfolios in units of the benchmark portfolio which leads to the fact that an equivalent measure is not needed. In the second approach the numéraire property of the Growth Optimal Portfolio is used. The numéraire portfolio defines an equivalent martingale measure and by change of measure using the Radon-Nikodým derivative, a real world pricing formula is derived which is the same as the one given by the first approach stated above.

A presente dissertação expõe o ambiente em que o Problema de Merton é construído e, baseando-se na bibliografia apresentada, constrói exemplos em softwares cujas especificidades podem colaborar na clareza da resolução. O software Matlab engloba as soluções numéricas, enquanto o software Maple é responsável pela solução de equações diferenciais ordinárias e parciais de forma simbólica. Apresenta-se modificações do Problema de Merton original como exercícios para melhor esclarecer os diferentes parâmetros abordados. Na seção final é apresentada a solução de viscosidade, uma alternativa quando a função valor não apresenta características desejáveis para a análise apresentada. / This dissertation explicit the environment which Merton’s problem is built, according to the presented bibliography, exemples are built in softwares whose specificity might help to clarify the solution. The Matlab software embraces numeric solutions, while Maple software is appropriate to solve ordinary and parcial differential equations in symbolic form. Some modifications are presented to Merton’s Problem as exercise to improve understanding on the variations adopted. On final section, viscosity solutions are presented as an alternative solution for when the value function does not possess the desirables properties that allow the analysis on focus.