OPOKU, E. Aplikovaná finanční optimalizace [online]. Brno: Vysoké učení technické v Brně. Fakulta strojního inženýrství. 2024.

Posudek vedoucího

Popela, Pavel

The author demonstrated that she had fulfilled the specified requirements pertaining to the theme of her diploma thesis. Furthermore, she has accomplished all of the primary objectives at the requisite level. The quantity of work completed by the author of the thesis is satisfactory, particularly in the areas of GAMS implementation and data treatment. The author conducted the majority of the work independently, initially attempting to resolve all issues without assistance. In the event of difficulties, she sought guidance from supervisor and subsequently engaged in reflective practice. In the course of her master studies, she drew upon the knowledge she had acquired in Brno through the Optimization 2 and Financial Mathematics courses. Furthermore, she has enhanced her knowledge base by engaging in independent study of relevant literature. In order to complete the requisite computations for her thesis, she undertook independent study of the GAMS software optimization package, and was subsequently able to utilize it successfully. In order to facilitate computational processes, she focused her attention on data sets pertaining to investment opportunities in her home country, Ghana. Initially, the author applied the Markowitz approach independently and successfully, implementing three parametrized models. Subsequently, she refined her initial approach and, with the guidance of her supervisor, processed the results for three implemented two-stage stochastic programs. The selected ideas, which included two stages, multiple time periods and suitable scenarios, were followed. However, the future use of the achieved results is limited. Nevertheless, they will undoubtedly assist the author in enhancing her expertise in Ghanaian commodities investment opportunities. As a supervisor, I have been encouraged by the author's efforts to initially develop some ideas related to two-stage programming, and particularly her constructive and prompt responses to my feedback. It is evident that further enhancements to the models and their calibration could be undertaken. Nevertheless, the author can take pride in the progress she has made since her arrival in Brno, where she has made a significant contribution to her knowledge in the field of financial optimisation. Her principal focus has been on data processing and implementation aspects. This also has a significant impact on the formal processing of the thesis, as some sections of the text appear to be underutilised or imbalanced in terms of the number of related pages. The author could devote more attention to establishing connections between the overview and the original sections, as well as to providing a more detailed presentation of the model descriptions in GAMS or through the use of mathematical formulas. It is therefore evident that the zip file containing all the GAMS programs is a more appropriate inclusion than initially anticipated. The graphical layout of the thesis is effective. Some typographical errors and other mistakes remained uncorrected. The author has demonstrated a satisfactory approach to the literature review. The evaluation by Theses tool is influenced by the fact that the author mostly cited fundamental resources in the overview sections, however, Theses also identified instances of secondary quotation from previous theses. In light of the aforementioned observations, it can be concluded that the author has achieved the desired outcomes of her thesis at a commendable level. Consequently, it is recommended that her thesis be presented at the thesis defence and assigned a grade of C.

Posudek oponenta

Hrabec, Dušan

The presented thesis deals with applying two known optimization models, the Markowitz Mean-Variance and Two-Stage Programming, when optimizing commodity investments. The thesis has 72 pages; however, the main body (from Introduction to Conclusion) is about 35 pages. The first half of the thesis is devoted to the introduction and theoretical framework description (mathematical programming and selected portfolio optimization models), while the second deals with applying and comparing two models mentioned above applied in the case study of Ghana for selected investment portfolios. The remaining 25 pages include appendices with data tables and GAMS models. Even if the theoretical part is well structured and describes the models used, it presents some material that has not been used further (e.g., integer programming) and involves some inaccuracies. For example, some of the figures do not properly describe the axes/units (the same with tables throughout the thesis); model (1.1) does not properly define the dimension of vector x; x* is used for both feasible and optimal solutions; the math notation of a set of real numbers is not unified; model (1.8) does not describe parameter Q; the source of Figure 1.3 is not referenced. On the other hand, the author has proven her abilities when studying and describing the portfolio optimization models. The models are then applied in the practical part of the thesis. The author chose data from Ghana and presented a case study of the selected models. The models were implemented in the GAMS optimization environment, and the results were successfully obtained. However, the abilities to present and discuss the results are clearly limited. Practical insights are poorly defined and presented. Overall, the author has successfully fulfilled the stated targets, but the practical part of the thesis could be better elaborated. Therefore, I recommend the thesis for the defense and suggest a grade of D.

eVSKP id 162422