Athens University Economics and Business (AUEB)
Prospect theory in portfolio management
Abstract
dc:descriptionThe present thesis focuses on the Prospect Theory (PT) of Kahneman and Tversky (1979), the most popular alternative to the classical Expected Utility Theory (EU). In this thesis I study all the basic elements of a Prospect Theory model: the (S-shaped) value function, the loss aversion, the probability distortion and their implications. I also analyze well-known puzzles in the literature such as the equity premium puzzle, the size and the value premium puzzle and I prove that these puzzles are real. I create new behavioral reward-risk models, always in a Prospect Theory framework, and I solve them in order to find the optimal allocations that ensure to Prospect Theory investors the maximum utility at each case. The first important behavioral factor in PT models is the value function. In my study I employ some of the most used value functional forms and I examine if these value functions are adequate to describe the PT investors’ behavior. I go one step further by constructing different type of value functions, employing reward and risk measures related to specific attitudes, in order to use them into behavioral models. Another essential factor that influences PT investors’ decisions is the loss aversion. In this thesis Ι explain how the loss aversion affects agents and Ι examine how the optimal solutions may change when agents are very, moderately, or less loss averse. Ι also incorporate in all my models probability distortion (which is often omitted in behavioral reward-risk models because of its complexity) and I examine how the introduction of the probability distortion may influence investors’ decisions and if this affection is independent by the used probability distortion functional form. I construct several behavioral reward-risk models using different value functional forms combining with several loss aversion indices and in each of these models I introduce probability distortion examining how the results change when objective and subjective probabilities are employed in the model. Finally, I present the results for each of these cases and I analyze how the including behavioral factors interact between each other. Another axis of my thesis is the examination of the equity, size, and value premium puzzles. I contribute to the literature by presenting results that differ from that of Benartzi and Thaler (1995) concerning the equity premium puzzle while I show how the preference of PT investors to positively skewed assets may explain the size and the value premium puzzles. Briefly, I provide new original evidence on the equity premium puzzle in several economic environments, under different economic conditions, varying time horizons where I incorporated many technical elements such as the changing risk aversion and probability distortion. I showed that my optimal solutions (optimal evaluation period of about 5-7 months) are in accordance to the Myopic Loss Aversion (MLA) hypothesis, but different from that of Benartzi and Thaler (annual optimal evaluation period). These solutions are not sensitive to the value function used in the model but they are sensitive to the period under consideration, to the economic situation at each period, the economic crises and to the probability distortion. This is another contribution of my work. Benartzi and Thaler support that the probability distortion is of second importance in such an analysis but I showed that the probability distortion does influence the optimal solutions by reducing their magnitude but without altering the MLA attitude as Blavatskyy and Pogrebna (2006) support. Concerning the size and the value premium puzzles, my results show that agents have a clear preference to the positively skewed small and value stock portfolios for all types of preferences instead of the negatively skewed large and the growth stock portfolios. According to their risk attitude they increase or decrease the proportion of their wealth that they invest to the small and the value stock portfolios but they always prefer these assets instead of the large and the growth stock portfolios. This preference does not change when I introduce probability distortion in the model.
Degree
thesis:*- Grantor dc:publisher
- Athens University Economics and Business (AUEB)
- Year dc:date
- 2015
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Zervoudi, Evanthia
- Ζερβούδη, Ευανθία
Subjects
dc:subject × 29- Θεωρία προοπτικής
- Διαχείρηση Χαρτοφυλακίου
- Στρέβλωση Πιθανοτήτων
- Χρηματοοικονομικός κίνδυνος, Μέτρηση και διαχείριση
- Θεωρία αποφάσεων
- Οικονομικά συμπεριφοράς
- Χρηματοοικονομική μοντελοποίηση
- χρηματοικονομικά puzzles
- Χρηματοοικονομική, συμπεριφορική
- Κίνδυνοι, Αποστροφή στους
- Μέτρα κινδύνου κι απόδοσης
- Prospect theory
- Portfolio management
- Probability Distortion
- Financial management
- Decision theory
- Financial modeling
- Behavioral economics
- Financial risk measurement and management
- financial puzzles
- equity premium puzzle
- size premium puzzle
- value premium puzzle
- loss aversion
- Risk measures and performance measures
- Κοινωνικές Επιστήμες
- Οικονομικά και Επιχειρήσεις
- Social Sciences
- Economics and Business
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/36107