|
|
Analysis of Various Approaches to Solving Optimization Tasks
Knoflíček, Jakub ; Samek, Jan (referee) ; Zbořil, František (advisor)
This paper deals with various approaches to solving optimization tasks. In prolog some examples from real life that show the application of optimization methods are given. Then term optimization task is defined and introducing of term fitness function which is common to all optimization methods follows. After that approaches by particle swarm optimization, ant colony optimization, simulated annealing, genetic algorithms and reinforcement learning are theoretically discussed. For testing we are using two discrete (multiple knapsack problem and set cover problem) and two continuous tasks (searching for global minimum of Ackley's and Rastrigin's function) which are presented in next chapter. Description of implementation details follows. For example description of solution representation or how current solutions are changed. Finally, results of measurements are presented. They show optimal settings for parameters of given optimization methods considering test tasks. In the end are given test tasks, which will be used for finding optimal settings of given approaches.
|
|
|
Portfolio Optimization in the German Stock Market
Bastin, Jan ; Musílek, Petr (advisor) ; Witzany, Jiří (referee) ; Budinský, Petr (referee)
The thesis focuses on the equity portfolio management with quantitative methods. We present 3 types of optimization objectives: One tries to find minimum variance portfolios, tangency portfolios and portfolios with maximized expected returns in the German stock market. It is possible to compare those investment opportunities with a market-cap weighted benchmark and an equal weighted portfolio. Expected returns of stocks are estimated with fundamental factor models. Risks of portfolios are estimated with 5 types of covariance matrices: the matrix calculated with historical returns, estimations with the single index model, Fama-French three factors model, fundamental factor model and the shrinkage method. Our results are doubled because of the demonstration of the impact of turnover constraint on portfolio performance measures (transaction costs are included in our calculations). One can see that optimized portfolios had attractive risk-return measures in the period 2005 - 2015. Benchmark and equal weighted portfolios were dominated and we consider them to be inefficient investments in our test.
|
|
|
Analysis of Various Approaches to Solving Optimization Tasks
Knoflíček, Jakub ; Samek, Jan (referee) ; Zbořil, František (advisor)
This paper deals with various approaches to solving optimization tasks. In prolog some examples from real life that show the application of optimization methods are given. Then term optimization task is defined and introducing of term fitness function which is common to all optimization methods follows. After that approaches by particle swarm optimization, ant colony optimization, simulated annealing, genetic algorithms and reinforcement learning are theoretically discussed. For testing we are using two discrete (multiple knapsack problem and set cover problem) and two continuous tasks (searching for global minimum of Ackley's and Rastrigin's function) which are presented in next chapter. Description of implementation details follows. For example description of solution representation or how current solutions are changed. Finally, results of measurements are presented. They show optimal settings for parameters of given optimization methods considering test tasks. In the end are given test tasks, which will be used for finding optimal settings of given approaches.
|
|
|
Using of AIMMS at solving of optimization models
Lacinová, Věra ; Jablonský, Josef (advisor) ; Zouhar, Jan (referee)
The bachelor's thesis is about AIMMS -- a system for mathematical modeling. The aim of the thesis is to create a simple user's guide for interested individuals that want to use this system on the basic level. Practical part is about making models of selected distribution problems. Thesis is organized into three main chapters. The first chapter is about linear programming and selected distribution problems of linear programming. There is outlined development of linear programming and its place in mathematical modeling. In this chapter, mathematical models of selected distribution problems as transportation problem, container-transportation problem, allocation problem, covering problem and assignment problem are formulated. The second chapter is about AIMMS and describes how to work with this system. It contains basic information about the system and its author J. Bisschop. The next part informs how to work with system AIMMS. There is a way of declaration of individual sections and procedures in Model Explorer discussed. The AIMMS Page Manager, a tool for creating end-user's pages, is described. By this tool the end-user can change data sets and obtain results of analyzed problems. The third chapter is practical part. It contains the description of selected distribution problems with the emphasis on their implementation in AIMMS. Appendix contains end-pages of selected distribution problems and the list of solvers supported by AIMMS.
|
|
| |
guest ::
