Advanced

Prospect Utility Portfolio Optimization

Lindeke, Niklas LU (2016) NEKN01 20161
Department of Economics
Abstract
Portfolio choice theory have in the last decades seen a rise in utilising
more advanced utility functions for finding optimal portfolios. This is
partly a consequence of the relatively simplistic nature of the quadratic
utility, which is often assumed in the classical mean-variance framework.
There have been some suggestions on how to find optimal portfolios in ac-
cordance to more realistic utility functions gathered from Prospect theory.
However, some of these methods suffer from practical drawbacks.
This paper proposes a method consisting of a mixture between two op-
timization techniques, in order to find a portfolio allocation that is optimal
in relation to the first four moments. In the empirical implementation, we
utilise... (More)
Portfolio choice theory have in the last decades seen a rise in utilising
more advanced utility functions for finding optimal portfolios. This is
partly a consequence of the relatively simplistic nature of the quadratic
utility, which is often assumed in the classical mean-variance framework.
There have been some suggestions on how to find optimal portfolios in ac-
cordance to more realistic utility functions gathered from Prospect theory.
However, some of these methods suffer from practical drawbacks.
This paper proposes a method consisting of a mixture between two op-
timization techniques, in order to find a portfolio allocation that is optimal
in relation to the first four moments. In the empirical implementation, we
utilise the S-shaped and Bilinear utility functions gathered from Prospect
Theory. Results hold in an in-sample testing environment. Improving ex-
pected utility, and the first three moments when tested against a standard
benchmark method, as well as in measurement of the Sharpe Ratio. (Less)
Popular Abstract
Portfolio choice theory have in the last decades seen a rise in utilising
more advanced utility functions for finding optimal portfolios. This is
partly a consequence of the relatively simplistic nature of the quadratic
utility, which is often assumed in the classical mean-variance framework.
There have been some suggestions on how to find optimal portfolios in ac-
cordance to more realistic utility functions gathered from Prospect theory.
However, some of these methods suffer from practical drawbacks.
This paper proposes a method consisting of a mixture between two op-
timization techniques, in order to find a portfolio allocation that is optimal
in relation to the first four moments. In the empirical implementation, we
utilise... (More)
Portfolio choice theory have in the last decades seen a rise in utilising
more advanced utility functions for finding optimal portfolios. This is
partly a consequence of the relatively simplistic nature of the quadratic
utility, which is often assumed in the classical mean-variance framework.
There have been some suggestions on how to find optimal portfolios in ac-
cordance to more realistic utility functions gathered from Prospect theory.
However, some of these methods suffer from practical drawbacks.
This paper proposes a method consisting of a mixture between two op-
timization techniques, in order to find a portfolio allocation that is optimal
in relation to the first four moments. In the empirical implementation, we
utilise the S-shaped and Bilinear utility functions gathered from Prospect
Theory. Results hold in an in-sample testing environment. Improving ex-
pected utility, and the first three moments when tested against a standard
benchmark method, as well as in measurement of the Sharpe Ratio. (Less)
Please use this url to cite or link to this publication:
author
Lindeke, Niklas LU
supervisor
organization
course
NEKN01 20161
year
type
H1 - Master's Degree (One Year)
subject
keywords
Utility Maximization, Portfolio Choice, Gradient Ascent, Sparse Group LASSO
language
English
id
8890111
date added to LUP
2016-09-09 14:04:56
date last changed
2016-09-09 14:04:56
@misc{8890111,
  abstract     = {Portfolio choice theory have in the last decades seen a rise in utilising
more advanced utility functions for finding optimal portfolios. This is
partly a consequence of the relatively simplistic nature of the quadratic
utility, which is often assumed in the classical mean-variance framework.
There have been some suggestions on how to find optimal portfolios in ac-
cordance to more realistic utility functions gathered from Prospect theory.
However, some of these methods suffer from practical drawbacks.
This paper proposes a method consisting of a mixture between two op-
timization techniques, in order to find a portfolio allocation that is optimal
in relation to the first four moments. In the empirical implementation, we
utilise the S-shaped and Bilinear utility functions gathered from Prospect
Theory. Results hold in an in-sample testing environment. Improving ex-
pected utility, and the first three moments when tested against a standard
benchmark method, as well as in measurement of the Sharpe Ratio.},
  author       = {Lindeke, Niklas},
  keyword      = {Utility Maximization,Portfolio Choice,Gradient Ascent,Sparse Group LASSO},
  language     = {eng},
  note         = {Student Paper},
  title        = {Prospect Utility Portfolio Optimization},
  year         = {2016},
}