Diese Dissertation beschäftigt sich mit mathematischen Modellen, die beschreiben, wie sich Menschen verhalten, wenn die Folgen ihrer Handlung im Moment der Entscheidung nur unvollständig bekannt oder gar gänzlich unbekannt sind: wie Menschen Unsicherheit wahrnehmen, wie sie dieser Unsicherheit gegenüber eingestellt sind und wie beides zusammen letztendlich ihre Entscheidungen bestimmt. In 3 empirischen Arbeiten wird unter jeweils unterschiedlicher Perspektive untersucht, wie sich die zentralen Konzepte "Nutzen" und "Erwartungen" messen lassen, welche der vielen Theorien menschliches Verhalten am besten beschreibt und welchen Erklärungsgehalt die Theorien insgesamt haben. Das erste Kapitel zeigt, wie schwierig die Messung und ökonometrische Schätzung von Präferenzparametern selbst in relativ einfachen Modellen für Entscheidungen unter Risiko sein kann. Kapitel zwei beschreibt die Ergebnisse mehrerer Experimente mit dem Ziel, die Eigenschaften subjektiver Erwartungen über Aktienmarktrenditen zu klären, und deren Erklärungsgehalt für tatsächliches Investitionsverhalten einzuschätzen. Das dritte und letzte Kapitel enthält einen experimentellen Test der Vorhersagen gängiger Theorien der Ambiguitätsaversion in Situationen, in denen sich Risiko und Unsicherheit vermischen. / This dissertation is concerned with mathematical models of choice under risk and uncertainty. Such models describe how people behave when the consequences of their actions are either unknown or unknowable at the point of decision: how they perceive that uncertainty, what their attitude towards it is and how the two jointly determine decisions. The dissertation shines a spotlight on such models from three different directions. All three chapters are empirical, designed to either study how the central theoretical constructs of "utility" and "beliefs" can be measured, which of the many theories describes choices best or what explanatory power the theories hold. The first chapter details how difficult the measurement and econometric estimation of preference parameters even in relatively simple models of choice under risk can be. The second describes the result of a series of experiments designed to probe the properties of subjective beliefs about stock market returns and their explanatory power for investment behavior. The third chapter, finally, offers an experimental test of the predictions of popular theories of ambiguity aversion in settings in which subjects face combinations of risk and uncertainty.

[pt] O presente estudo busca analisar o comportamento do investidor brasileiro em diferentes ambientes de taxas de juros e compara esse comportamento com a teoria tradicional de finanças e as finanças comportamentais. Além de dados quantitativos disponíveis do mercado de capitais brasileiro, o estudo desenvolve uma pesquisa qualitativa, observacional de campo – de autoria dos autores. A pesquisa de campo busca adaptar para o mercado brasileiro o trabalho de Yoav Ganzach and Avi Wohl (2018) A Behavioral Theory of the Effect of the Risk-Free Rate on the Demand for Risky Assets que sugere que taxas de juros mais baixas mudam a propensão a tomar risco por parte dos investidores. Esse tipo de abordagem vai de encontro ao que sugere a microeconomia clássica, segundo a qual, o que importa é o prêmio de risco, ou o retorno superior ao retorno sem risco. Dito de outra forma, segundo a literatura clássica, o nível de retorno livre de risco não importa. O presente trabalho reproduz experimentos no mercado brasileiro visando compreender se prêmio de risco é uma variável relevante ou se o nível de juros altera a forma de se investir. Além disso o trabalho busca entender se há distinção entre investidores de varejo e de alta renda. / [en] The present study analyses the behaviour of brazilian investors in different interest rate environments and compares this behaviour with traditional theory of finance and behavioural finance. In addition to quantitative data available from the Brazilian capital market, the study develops a qualitative, observational field research – authored by the authors. The field research seeks to adapt the work of Yoav Ganzach and Avi Wohl (2018) A Behavioral Theory of the Effect of the Risk-Free Rate on the Demand for Risky Assets that suggests that lower interest rates change investors propensity to take risk. This type of approach goes against what the classical microeconomy suggests, according to which, what matters is the risk premium, or the higher return to risk-free return. Put another way, according to classical literature, the level of risk-free return does not matter. The present work reproduces experiments in the Brazilian market to understand whether risk premium is a relevant variable or if the level of interest changes the way of investing. In addition, the paper seeks to understand whether there is a distinction between retail and high-income investors.