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.

The principle of equal treatment and non-discrimination in labor relations Abstract This dissertation focuses on the analysis of the legal regulation of the Czech Republic containing the principle of equal treatment of employees and the prohibition of their discrimination in labor relations. The other objective is to compare the legal regulations of other European Union Member States and to apply the principle of equal treatment and non- discrimination in practice, including the abundant case law of The Court of Justice of the European Union, the European Court of Human Rights and the less case law of the Czech courts. The content of this dissertation above all is an introduction of the principle of equal treatment and non- discrimination, differentiation of material and formal equality including specification of individual discriminatory reasons and an introduction of the most important legal sources in the international legislation, standards of the European anti-discriminatory law and domestic legal adjustment. Another part describes the rules of equal treatment and the ban of discrimination applied in real life in the area of remuneration in terms of rewarding somebody the same amount for the work of same value, and a possibility of achieving a functional or other types of promotion without...

Social learning is a topic of many studies. We tested the effect of social learning on the acquisition of avoidance against aposematic prey. We have chosen wild-caught adult and naive hand-reared juvenile great tits (Parus major) as a model predator species, because their individual avoidance learnig of aposematic prey is well-studied. We used red and green paper dummies of bugs with a mealworm (Tenebrio molitor) stuck underneath as an artificial prey. Mealworms were soaked in water or in bitter-tasting solution of quinine. We used two types of conspecific demonstrators - naive birds that showed aversive reactions while tasting an unpalatable prey and experienced birds that were trained not to handle the unpalatable pray at all. We compared effects of both demonstrators on discrimination learning and we tested whether these effects differ in adult and juvenile birds. Observing an experienced demonstrator had an effect on the performance of observers at the beginning of learning process. The observers did not reject the unpalatable prey completely, but the number of trials in which they correctly chose the palatable prey was higher in comparison with birds that observed naive demonstrators and birds from the control group with no demonstrator. Latencies to handle the unpalatable prey in the first...

Social learning in animals is a frequently studied topic, including birds. Great tits belong to frequently tested passerine species in this context. For this experiment we have chosen another tit species as model birds - adult and juvenile blue tits (Cyanistes caeruleus) and coal tits (Periparus ater). We tested an effect of social information on discrimination learning between palatable and unpalatable prey. Red and green artificial prey items resembling shieldbugs with a mealworm glued underneath were offered to birds. One variant was palatable, the other was unpalatable, soaked in bitter substance. Experimental groups were allowed to observe a pretrained tutor of the same species choosing palatable prey variant and rejecting the unpalatable one. The birds were then subjected to an individual discrimination learning task with simultaneously offered palatable and unpalatable prey items. The control group was not allowed to observe tutor and learned only individually. On the following day, all birds participated in a memory test. We tested if social information has an influence on avoidance learning in juvenile and adult birds of both species. We found that both species of tits were similarly successful in discrimination learning and there was no difference in success in performance between adult...

This master thesis deals with bankruptcy prediction models for construction companies doing business in Czech Republic. Terms important for understanding the issue are defined in the theoretical part. In analytical part, there are five current bankruptcy prediction models tested on the analysed sample and resulted accuracy compared with original ones. On the basis of knowledges acquired, there is developed a brand-new bankruptcy prediction model.

The purpose of this doctoral thesis is to create a new bankruptcy prediction model and also to design how to use this model for the purposes of credit risk measuring. The starting-point of this work is the analysis of traditional bankruptcy models. It was found out that the traditional bankruptcy model are not enough effective in the current economic conditions and it is necessary to create a new ones. Based on the identified deficiencies of the traditional models a set of two new model series was created. The first series of the created models is based on the use of parametric methods, and the second one is based on the use of newer nonparametric approach. Moreover, a set of factors which are able to identify an imminent bankruptcy was analyzed. It was found, that significant signs of imminent bankruptcy can be identified even five years before the bankruptcy occurs. Based on these findings a new model was created. This model incorporates variables of static and even dynamic character for bankruptcy prediction purposes. The overall classification accuracy of this model is 92.27% of correctly classified active companies and 95.65% of correctly classified bankrupt companies.

The European green lizard, Lacerta viridis, is in the Bohemia region stated as critically endangered species. Populations in this region are located beyond the northern border of continuous range of this species and are closely related to the "riverine phenomenon", and deeply engorged river valleys. The chosen locality in Tiché údolí is a subject to a long-term conservational management aimed to strengthen and maintain abundance of the local population. This management is a direct output of a previous study of this population performed in years 1995-1997. Main goal of the presented study is to compare current population characteristics with the older study. The locality has been visited 119 times in years from 2011 to 2014. The studied population now displays higher abundance and inhabits a larger area. The author also performed a spatial analysis of the places with presence of an observed individual in order to determine and evaluate significance of the chosen abiotic factors for habitat discrimination. The results indicate that positive discrimination is based on the presence of a rock debris and a hiding place. Strongest factors towards negative discrimination were high percentages of grass and high vegetation coverage. This study also contains and discusses ecological, ethological and...

Thesis deals with complex process of creation of new bankruptcy model for predicting business failure, while this process involves selection of quality sample, verification of classification accuracy of already existing bankruptcy models, profile analysis and finally the derivation of specific equation of bankruptcy model. The derivation is performed by using two selected statistical methods, discriminant analysis and logistic regression. Two bankruptcy models Bürger's index DA12 and Bürger's index LR12 were derived by using the mentioned statistical methods. The new models distinct advantage is, unlike already existing and renowned bankruptcy models, that they are focused on classification of micro and small enterprises in terms of Czech Republic, while classification accuracy one year before failure is by individual models 74.8 % and 81.87 %. Derived models have clear interpretation (no grey zone) and easy calculation, which brings a possibility for micro and small entrepreneurs to check their business partners in terms of failure prediction.

The theory of redistribution systems is a practical extension of a game theory, which deals with a redistribution within a social system of more than two players with di?erent performances and ability to create coalitions. This thesis is divided into three chapters. The ?rst chapter describes the known knowledge of a game theory. The second chapter deals with the theory of redistribution systems. Using an elementary redistribution system and its generalization group behaviour when dividing a payment, achieved by a collective performance, is described. This part introduces the extension of the redistribution system to a compound redistribution system with a fractal structure. Furthermore the theory of discriminatory equilibrium and the theory of commonly acceptable equilibrium are veri?ed using the elementary redistribution system and utility theory. The third chapter deals with an application to the allocation of funds among faculty departments. A game theory approach was used to reduce the game to a non-cooperative game of two players by using the forming of coalitions. Also the theory of redistribution systems was applied when a reduction was used to create a non-cooperative two-player game. This reduced non-cooperative game between two players was converted to a cooperative play of more than two players by changing the rules of the game and allowing a formation of coalitions. In the practical part both of these approaches are compared with real data and a current state.

In this paper, we propose techniques for adaptation of speaker recognition systems. The aim of this work is to create adaptation for Probabilistic Linear Discriminant Analysis. Special attention is given to unsupervised adaptation. Our test shows appropriate clustering techniques for speaker estimation of the identity and estimation of the number of speakers in adaptation dataset. For the test, we are using NIST and Switchboard corpora.