Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.

  

Portfolio Optimization for Small Investors

Author(s):

B. Liu
Conference/Journal:

Semester Thesis, FS15 (10431)
Abstract:

It is often hard to compose an optimal portfolio out of an huge palette of stocks. It requires sound standing knowledges about the stock market, the general economic en- vironment and the related companies. However, Markowitz has proposed in 1952 that the relationship between mean return and variance of return could be used to formulate the rule which an investor would follow when composing an optimal portfolio[1]. In his "Portfolio Selection" he has provided a systematic way to approach the solution of the portfolio optimization problem. This project is build on solving and testing this kind of optimization problems. It is especially out of interest of the small investors, because the solve-and-test toolbox which is the main achievement of this project, provides, at first, an user-friendly interface which quantizes the user preferences in an easy and understandable way. Second, it modularizes different market participants such like banks and investors and other function blocks like optimization strategies. These modulus are allowed to be exchanged or shared and new module could be introduced. Third, the data retrieving, data handling and back-testing procedures are automated. The user could concentrate on essential issues such like modelling or risk measurement techniques rather than dealing with programming technicalities. Beside the back-testing framework four optimization models are implemented. They are the Mean-Variance model[1], the Mean-Absolute- Deviation (MAD) model[2], the Minimax model[3] and the Conditional-Value-at-Risk (CVaR) model[4]. Additional to their basic constraints, a linear rebalancing constraint is introduced. This constraint includes a piecewise-constant transaction cost structure and couples the previous portfolio into the model[5][6][7]. With the rebalancing constraint the portfolio history shows a more smoothed development rather than drastically chang- ing in portfolio positions. Furthermore it shows that with the framework that has been implemented, the portfolio selection process is enhanced and it's analysis is simplified.

Supervisors: Robin Vujanic, Angelos Georghiou, John Lygeros

Year:

2015
Type of Publication:

(13)Semester/Bachelor Thesis
Supervisor:

J. Lygeros

File Download:

Request a copy of this publication.
(Uses JavaScript)
% Autogenerated BibTeX entry
@PhdThesis { Xxx:2015:IFA_5193
}
Permanent link