|
|
Methods of longitudinal data analysis
Jindrová, Linda ; Volf, Petr (advisor) ; Prášková, Zuzana (referee)
Práce se zabývá longitudinálními daty - měřeními, která jsou prová- děna opakovaně na stejných subjektech. Popisuje r·zné typy model·, které jsou vhodné pro jejich analýzu. Postupuje od nejjednodušších lineárních model· s pevnými nebo náhodnými efekty, přes lineární a nelineární modely se smíšenými efekty, až ke zobecněným lineárním model·m a generalized estimating equati- ons (GEE). Vždy je uveden tvar modelu a zp·sob odhadu parametr·. Jednotlivé modely jsou také porovnávány mezi sebou. Teoretické poznatky jsou doplněny aplikacemi na reálná data. Pomocí lineárních model· analyzujeme data o výrobě v USA, nelineární modely využijeme k vysvětlení závislosti koncentrace léčiva v krvi na čase a GEE aplikujeme na data týkající se dýchacích potíží u dětí. 1
|
|
|
Classification based on longitudinal observations
Bandas, Lukáš ; Komárek, Arnošt (advisor) ; Kulich, Michal (referee)
The concern of this thesis is to discuss classification of different objects based on longitudinal observations. In the first instance the reader is introduced to a linear mixed-effects model which is useful for longitudinal data modeling. Description of discriminant analysis methods follows. These methods ares usually used for classification based on longitudinal observations. Individual methods are introduced in the theoretic aspect. Random effects approach is generalized to continuous time. Subsequently the methods and features of the linear mixed-effects model are applied to real data. Finally features of the methods are studied with help of simulations.
|
|
|
Multi-level quantitative evaluation of the use of foreigners in Czech football
Riedl, Jakub ; Crossan, William Morea (advisor) ; Pecha, Ondřej (referee)
Title: Multi-level quantitative evaluation of the use of foreigners in Czech football Objectives: The purpose of this thesis is to evaluate how foreign soccer players influence the Czech first league football in seasons from 1993/1994 to 2019/2020, and use multilevel modeling to analyze longitudinal data to find answers to these questions: 1. Do foreign players effect attendance in Czech first soccer league? 2. Do foreigners effect the number of Czech players in the highest soccer league? A secondary goal is to find out if multilevel analysis is a suitable method to evaluate sport migration in a primary sport in a semi-periphery country. Methods: In the master's thesis multilevel analysis with longitudinal data is used to explain dependent variables which were acquired from the Czech first football league between the seasons 1993/94 and 2019/20. Results: The results of this work show that foreign players do not have an effect on attendance because the results were statistically insignificant. The number of foreign players in the Czech league is increasing on average by 0,22 players per year in one club. On the other hand, Czech players were decreasing in all 27 seasons by 0,13 per year per club. The relationship between the dependent variable of Czech players and independent variable of foreign...
|
|
|
Methods of longitudinal data analysis
Jindrová, Linda ; Volf, Petr (advisor) ; Prášková, Zuzana (referee)
Práce se zabývá longitudinálními daty - měřeními, která jsou prová- děna opakovaně na stejných subjektech. Popisuje r·zné typy model·, které jsou vhodné pro jejich analýzu. Postupuje od nejjednodušších lineárních model· s pevnými nebo náhodnými efekty, přes lineární a nelineární modely se smíšenými efekty, až ke zobecněným lineárním model·m a generalized estimating equati- ons (GEE). Vždy je uveden tvar modelu a zp·sob odhadu parametr·. Jednotlivé modely jsou také porovnávány mezi sebou. Teoretické poznatky jsou doplněny aplikacemi na reálná data. Pomocí lineárních model· analyzujeme data o výrobě v USA, nelineární modely využijeme k vysvětlení závislosti koncentrace léčiva v krvi na čase a GEE aplikujeme na data týkající se dýchacích potíží u dětí. 1
|
|
|
Classification based on longitudinal observations
Bandas, Lukáš ; Komárek, Arnošt (advisor) ; Kulich, Michal (referee)
The concern of this thesis is to discuss classification of different objects based on longitudinal observations. In the first instance the reader is introduced to a linear mixed-effects model which is useful for longitudinal data modeling. Description of discriminant analysis methods follows. These methods ares usually used for classification based on longitudinal observations. Individual methods are introduced in the theoretic aspect. Random effects approach is generalized to continuous time. Subsequently the methods and features of the linear mixed-effects model are applied to real data. Finally features of the methods are studied with help of simulations.
|
guest ::
