This thesis investigates the choices under risk in the framework of non-expected utility theories. One of the key contributions of this thesis is providing an approach that allows for a complete characterisation of Cumulative Prospect Theory (CPT) preferences without prior knowledge of the reference point. The location of the reference point that separates gains from losses is derived endogenously, thus, without any additional assumptions on the decision maker’s risk behaviour. This is different to the convention used in the literature, according to which, the reference point is preselected. The problem arising from imposing the location of the reference point is that the underlying preference conditions might not be alligned with the predictions made by the model. Consequently, it is difficult to verify such a model or to test it empirically. The present contribution offers a set of normatively and descriptively appealing preference conditions, which enable the elicitation of the reference point from the decision maker’s behaviour. Since these conditions are derived using objective probabilities, they can also be applied to settings such as health or insurance, where the continuity of the utility function is not required. As a result, the obtained representation theorem is not only the most general foundation for CPT currently available, but it also provides further support for the use of CPT as a modelling tool in decision theory and fi…nance. Another contribution that this thesis can be credited with is an application of rank-dependent utility theory (RDU) to the problem of insurance demand in the monopoly market affected by adverse selection. The present approach extends the classical model of Stiglitz (1977) by accounting for an additional component of heterogeneity among consumers, the heterogeneity in risk perception. Speci…fically, consumers employ distinctive probability weighting functions to assess the likelihood of risky events. This aspect of consumers’' behaviour highlights the importance that the probabilistic risk attitudes within the RDU framework, such as optimism and pessimism, have for the choice of insurance contract. The analysis yields a separating equilibrium, with full insurance for a sufficiently pessimistic decision maker. An important implication of this result is that any low-risk individual who sufficiently overestimates his probability of loss will induce the uninformed insurer to o¤er him full coverage, thereby, affecting the high-risk type adversely. This outcome is consistent with the recent empirical puzzle regarding the correlation between ex-post risk and insurance coverage, according to which, agents with low exposure to risk receive a larger amount of compensation. By providing an explanation of this pattern of individual behaviour, the current work demonstrates that theory and practice of insurance demand can be reconciled to a greater extent. The paper also provides a behavioural rationale for policy intervention in the market with RDU agents, where the initial distortions in contracts due to unobservable risks are aggravated by the non-linear weighting of probability of a risky event.

[pt] Baseado na Teoria Cumulativa da Perspectiva, três ensaios são apresentados nessa tese. Todos os três trabalhos estão conectados pelo entendimento aprofundado da Função de Ponderação de Probabilidade e suas conexões cenários de decisão sob risco. O primeiro ensaio é um trabalho empírico utilizando a teoria da perspectiva para analisar o viés do efeito de enquadramento em decisões de investimentos em certos países emergentes: Brasil, China, Russia, México e África do Sul. Em todos os casos, identificamos empiricamente o poder preditivo da teoria da perspectiva para os retornos dos ativos. Também encontramos que a função de ponderação de probabilidade é o fator mais importante para o poder preditivo. O segundo ensaio é um trbalho teórico propondo uma axiomatização da função de ponderação de Goldstein-Einhorn. Desde 1987, a conhecida função de ponderação de Goldstein-Einhorn é largamente utilizada em trabalhos em muitos artigos empíricos e teóricos. Richard Gonzalez e GeorgeWu propuseram uma axiomatização para esta função em 1999. O trabalho que apresentamos analisa a condição de preferência dos autores e encontra uma família maior de funções de ponderação. Fornecemos exemplos úteis e sugerimos uma nova condição de preferência que é necessária e suficiente para a função de Goldstein- Einhorn. Esta nova condição de prefer6encia simula o comportamento das pessoas em situações que envolvem atitutes arriscadas. O terceiro ensaio propõe uma medida para as características psicológicas chamadas de atratividade e discriminabilidade, no contexto das funções de ponderação de probabilidades. Esse conceitos são importantes para nos ajudar a entender como algumas emoções influenciam nosso comportamento. Propomos medidas no sentido absoluto e relativo e as comparamos com alguns exemplos particulares encontrados na literatura. Nossos resultados são consistentes com o entendimento qualitativo encontrado na literatura e fornece um entendimento quantitativo para ele. / [en] Based on Cumulative Prospect Theory, three essays are presented in this thesis. All three works are linked by a deeper understanding of Probability Weighting Functions and its connection with decisions in a risk scenario. The first essay is an empirical work using prospect theory to analyze the narrow framing bias in investment decisions in certain emerging countries: Brazil, China, Russia, Mexico and South Africa. In all cases, we empirically identified the predictive power of prospect theory for stock returns. We also found that the probability weighting function is the most important factor in this predictive power. The second essay is a theoretical work proposing an axiomatization for the Goldstein-Einhorn weighting function. Since 1987, the well known Goldstein- Einhorn Weighting Function is widely used in many empirical and theoretical papers. Richard Gonzalez and George Wu proposed an axiomatization for it in 1999. The present work analyses their preference condition and finds a bigger family of weighting functions. We provide useful examples and suggest a new preference condition which is necessary and sufficient for Goldstein-Einhorn function. This new preference condition simulates the behavior of people in risky attitudes. The third essay propose a measure to evaluate the psychological features of attractiveness and discriminability in the context of probability weighting functions. These concepts are important to help us understand how some emotions drive our behavior. We propose measures in absolute and in the relative sense and compare with some particular cases found in the literature. Our findings are consistent with the qualitative understanding widespread in the literature and provide a quantitative analysis for it.

Made available in DSpace on 2010-04-20T21:00:18Z (GMT). No. of bitstreams: 3 martincasalsturma2003.pdf.jpg: 11517 bytes, checksum: 930c75c74ff8269d877983110854646d (MD5) martincasalsturma2003.pdf: 975293 bytes, checksum: f41cfd3a5f3e659d07e2861acf4e23d4 (MD5) martincasalsturma2003.pdf.txt: 107784 bytes, checksum: e96ad0fb809a12f13ba7d3f08e24c1ee (MD5) Previous issue date: 2006-02-15T00:00:00Z / O objetivo deste trabalho é analisar a alocação de investimentos no mercado acionário brasileiro, utilizando a teoria do prospecto de Tversky e Kahneman (1979) e o conceito de Aversão a Perdas Míope (Myopic Loss Aversion) proposto por Benartzi e Thaler (1995). Foram levantados através de experimento de laboratório os parâmetros da função de valor e da função de ponderação de probabilidades da teoria do prospecto e foi verificada a alocação de investimentos entre ações e renda fixa que maximizam a utilidade. Chegamos à conclusão que o total de recursos atualmente direcionados ao mercado de ações no Brasil, que é de aproximadamente 2,7% para pessoas físicas e de 6,0% para pessoas jurídicas, é compatível com a teoria do prospecto. / The objective of this study is to analyze the investment allocation in the Brazilian stock market, using Tversky and Kahneman’s prospect theory (1979) and the concept of myopic loss aversion proposed by Benartzi and Thaler (1995). We run a laboratory experiment to obtain the parameters of the value function and the probability weighting function of the prospect theory and identify the allocation that maximizes utility in the Brazilian Market We conclude that the actual allocation of investment in the stock market, of around 2.7% for individuals and around 6% for all the segments, is in accordance with the prospect theory.