|
Vliv hazardu na kriminalitu: evidence z České Republiky
Lupač, Milan ; Dušek, Libor (advisor) ; Špecián, Petr (referee)
The focus of this thesis is to examine the relationship between gambling and crime in the Czech environment, where gambling is broadly available. Data about the individual gambling machines and tables together with the data about offenses in particular police districts were used in order to estimate the effect of gambling on crime. The final dataset observes 388 geographical units over the life span between April 2013 and December 2015. The study employs three estimation techniques the OLS, Poisson regression and Negative binomial regression to estimate the effect of gambling on crime. The main variable representing the size of gambling is the number of slot machines as these are the most broadly available type of gambling. The final estimated relationship between crime and slot machines is that one additional slot machine is associated with an increase in crime by 0.3-0.5% depending on the method and frequency. On the contrary, the effect of casino games, electromechanical roulettes, and dice devices on crime was found to be statistically insignificant. In addition, the study also analyses particular types of crimes, finding that gambling has an impact particularly on crimes that involve material benefits as opposed to the violent crimes. Moreover, it also conducts a what-if analysis demonstrating the estimated impact of reduction of gambling on the substantial drop of the number of offenses over the observed period was rather limited and account for 937 offenses.
|
|
Comparison of approaches to creating credit scoring models
Hofman, Elena ; Šedivý, Jan (advisor)
This work is focused on the management of a credit risk related to the traditional bank lending business to individuals. The paper deals with a theory of measuring risk with help of PD (Probability of Default) parameter when different scoring models are used. The goal is to outline an issue with the credit risk and its management in general, attention is paid to details of a process of creating scoring models. There are three specific modeling techniques listed, namely logistic regression, decision trees and neural networks. Methods are explained in detail and are given possibilities of mutual comparison. The application part is devoted to the evaluation and comparison of credit scoring models based on these methods.
|
|
Trh hráčů americké Národní fotbalové ligy: Jsou hráčům stále vypláceny mzdy odpovídající tržní situaci monopsonu?
Miškovský, Karel ; Hudík, Marek (advisor) ; Koubek, Ivo (referee)
The main goal of this thesis is to find out whether National football league players are, even 15 years after the birth of unrestricted free agency, still paid monopsonistic salaries or whether competition among NFL teams eliminated them. After a theoretical discussion, which will help to form expectations about the player market, models in line with the standard theory of labor compensation and with Becker's human capital theory are estimated. The part of the research following the standard theory focuses mainly on estimations of players' marginal revenue product and subsequently on comparison of their wages with their MRP. To do so, OLS regressions, as well as quantile regressions, are run. The part of the research following the human capital theory has a supporting role and is used to further confirm the previous findings. It is represented by OLS player salary estimation. The hypothesis that players under a full control of their teams are still paid salaries below their MRP cannot be rejected, thus confirming a presence of monopsonistic salaries. A significant effect of free agency status on player salaries is also found. Exclusive rights players are paid significantly lower salaries than all free agents, while restricted free agents are paid significantly lower salaries than unrestricted free agents. Superstar players are found to have salaries in excess of their MRP regardless of their free agency status.
|
|
Co-Learning in Cartesian Genetic Programming
Korgo, Jakub ; Grochol, David (referee) ; Wiglasz, Michal (advisor)
This thesis deals with the integration of co-learning into cartesian genetic programming. The task of symbolic regression was already solved by cartesian genetic programming, but this method is not perfect yet. It is relatively slow and for certain tasks it tends not to find the desired result. However with co-learning we can enhance some of these attributes. In this project we introduce a genotype plasticity, which is based on Baldwins effect. This approach allows us to change the phenotype of an individual while generation is running. Co-learning algorithms were tested on five different symbolic regression tasks. The best enhancement delivered in experiments by co-learning was that the speed of finding a result was 15 times faster compared to the algorithm without co-learning.
|
|
Sustainable Governance of the Visegrad Countries
Ivantsiv, Olena ; Vykoukal, Jiří (advisor) ; Handl, Vladimír (referee)
With their accession to the EU the Visegrad countries subscribed to the fundamental objective of the Union under the Lisbon Treaty - sustainable development. They have undergone substantial reforms, brought their policies into compliance with EU standards and regulations. Nevertheless, a lot of work should still be done in the Visegrad Four in order integrate sustainable development approach into all of the fields of political activity and reorganize their decision-making models according to the new challenges. This study constitutes an analysis of the Visegrad states' performance in ensuring sustainable governance in the period 2005-2010. The research is based on the two editions of Sustainable Governance Indicators, developed by the Bertelsmann Stiftung, and published in 2009 (period of review: January 2005 - March 2007) and in 2011 (period of review: May 2008 - April 2010). In order to assess sustainability of the four Visegrad democracies the study provides a comprehensive comparative analysis of these states' performance and retraces the dynamics of their progress/regress in terms of ensuring sustainability. It also explores the main tendencies of the Visegrad region's development regardless of particular country, identifies the main strengths and weaknesses of the region in terms of...
|
|
Comparison of selected classification methods for multivariate data
Stecenková, Marina ; Řezanková, Hana (advisor) ; Berka, Petr (referee)
The aim of this thesis is comparison of selected classification methods which are logistic regression (binary and multinominal), multilayer perceptron and classification trees, CHAID and CRT. The first part is reminiscent of the theoretical basis of these methods and explains the nature of parameters of the models. The next section applies the above classification methods to the six data sets and then compares the outputs of these methods. Particular emphasis is placed on the discriminatory power rating models, which a separate chapter is devoted to. Rating discriminatory power of the model is based on the overall accuracy, F-measure and size of the area under the ROC curve. The benefit of this work is not only a comparison of selected classification methods based on statistical models evaluating discriminatory power, but also an overview of the strengths and weaknesses of each method.
|
|
Hypotéza endogenity teorie OCA v zemích střední a východní Evropy
Míč, Zdeněk
This diploma thesis examines the endogeneity hypothesis of the optimum currency area theory. Cyclical components of gross domestic product and index of industrial production are identified by three selected techniques. The convergence of the surveyed countries is assessed by correlation and cluster analysis. Correlation coefficients between the euro area and the 26 countries are used as dependent variables. Independent variables represent the optimum currency area criteria. Influence of the optimum currency area criteria upon the correlation of business cycles is verified in the 66 simple and 11 multivariate regressions. According to the empirical results, convergence of business cycles is endogenous, endogeneity hypothesis is not rejected.
|
|
A point process driven by a Gaussian field
Scheib, Karel ; Beneš, Viktor (advisor) ; Šedivý, Ondřej (referee)
The thesis investigates the search for dimension reduction subspace for the Poisson point process driven by a Gaussian random eld. The work describes the method called sliced inverse regression, which is applied to a point process driven by random eld. Its functionality in mentioned context is then proved. This method is in several ways implemented and tested in R software environment on random data. The individual implementations are described and results are then compared with each other.
|
|
Spatial analysis of illegal migration in Czechia 2005-2007
Mahová, Eva ; Štych, Přemysl (advisor) ; Schneider, Michal (referee)
Spatial analysis of illegal migration in the Czech Republic 2005-2007 Abstract This thesis deals with the spatial analysis of illegal migration across so-called green border from the Czech Republic to Germany and Austria in 2005-2007. The main data set was acquired from Czech Alien Police Service and Ministry of the Interior of the Czech Republic and contains also spatial component. For all places of detention various indicators characterizing geomorphology, landscape structure or distances from the closest routes, settlements and border checkpoints are computed. We can identify main aspects related to the choice to illegally cross state borders by merging the database of anonymous personal data of persons detained while illegally crossing the border and geographical data describing the nature of the place of detention. Keywords: illegal migration, GIS, Python, spatial analysis, correlation, regression
|
|
Genetic programming - Java implementation
Tomaštík, Marek ; Kuba,, Martin (referee) ; Matoušek, Radomil (advisor)
This Master´s thesis implements computer program in Java, useful for automatic model generating, specially in symbolic regression problem. Thesis includes short description of genetic programming (GP) and own implementation with advanced GP operands (non-destructive operations, elitism, exptression reduction). Mathematical model is generating by symbolic regression, exacly for choosen data set. For functioning check are used test tasks. Optimal settings is found for choosen GP parameters.
|