SLÁVIK, Ľ. Dynamická faktorová analýza časových řad [online]. Brno: Vysoké učení technické v Brně. Fakulta strojního inženýrství. 2021.
The thesis of Bc. Slávik aims at a representation of the time series by dynamic factors and its application in clustering. This novel method of time series clustering is used in the analysis of fire weather index recordings at selected stations in British Columbia in 2003. The author also suggests utilizing the results of the dynamic factor analysis in outliers identification. The stability of the clusters is examined with respect to the outliers as well as the studied month intervals. Bc. Slávik worked independently, he was trying to understand the applied methods carefully. To achieve this, he had to study a wide range of sources, which were transferred into a comprehensible text. Moreover, the real data analysis required a lot of coding in R. I'm also happy to add, that Bc. Slávik won third place with this thesis in the Competition of university students (SVOČ 2021) in the section Theory of probability, mathematical statistics, econometry, and financial mathematics. It was my pleasure to supervise the thesis of Bc. Slávika. Moreover, I believe, this thesis will be a benefit for discussion with the members of the Canadian Forest Service about the clustering of stations with similar trends for optimal forest management system in British Columbia.
| Kritérium | Známka | Body | Slovní hodnocení |
|---|---|---|---|
| Splnění požadavků a cílů zadání | A | ||
| Postup a rozsah řešení, adekvátnost použitých metod | A | ||
| Vlastní přínos a originalita | A | ||
| Schopnost interpretovat dosažené výsledky a vyvozovat z nich závěry | A | ||
| Využitelnost výsledků v praxi nebo teorii | A | ||
| Logické uspořádání práce a formální náležitosti | A | ||
| Grafická, stylistická úprava a pravopis | A | ||
| Práce s literaturou včetně citací | A | ||
| Samostatnost studenta při zpracování tématu | A |
The candidate met the stated objectives of the thesis, introduced a dynamic factor model, described an algorithm for maximum likelihood parameter estimation and implemented it in the R environment using the EM algorithm. He added cluster analysis methods to classify the studied time series into clusters. He applied the theoretical results to model the Fire Weather Index time series obtained from measurements at weather stations in British Columbia. The results obtained are of practical use in assessing fire hazards in the study areas. The graphical representation of zones by fire danger level is very valuable. I consider the thesis very successful, the candidate has shown that he is able to handle the difficult theoretical parts well, to implement them by computer and to obtain new results that have immediate practical application. I propose a grade A and strictly recommend the thesis for defence.
| Kritérium | Známka | Body | Slovní hodnocení |
|---|---|---|---|
| Splnění požadavků a cílů zadání | A | ||
| Postup a rozsah řešení, adekvátnost použitých metod | A | ||
| Vlastní přínos a originalita | A | ||
| Schopnost interpretovat dosaž. výsledky a vyvozovat z nich závěry | A | ||
| Využitelnost výsledků v praxi nebo teorii | A | ||
| Logické uspořádání práce a formální náležitosti | B | ||
| Grafická, stylistická úprava a pravopis | B | ||
| Práce s literaturou včetně citací | A |
eVSKP id 133320