|
|
Models of binary time series
Kunayová, Monika ; Zichová, Jitka (advisor) ; Cipra, Tomáš (referee)
This bachelor thesis deals with the time series of binary variables that exist in many social spheres. The indicator may denote a certain value being exceeded or a phenomenon occurring. We study a model of logistic autoregression and its properties, partial likelihood function which allows us to work with dependent data, and derive useful relationships for a practical application that consists of time series simulation and real data analysis using free software R.
|
|
|
Comparison of Heuristic and Conventional Statistical Methods in Data Mining
Bitara, Matúš ; Žák, Libor (referee) ; Bednář, Josef (advisor)
The thesis deals with the comparison of conventional and heuristic methods in data mining used for binary classification. In the theoretical part, four different models are described. Model classification is demonstrated on simple examples. In the practical part, models are compared on real data. This part also consists of data cleaning, outliers removal, two different transformations and dimension reduction. In the last part methods used to quality testing of models are described.
|
|
|
Risk factors affecting endometriosis in women of reproductive age, ALSWH Study
Olšarová, Karolína ; Dzúrová, Dagmar (advisor) ; Pikhart, Hynek (referee)
This thesis adapts a life course approach in epidemiology to endometriosis. Endometriosis is a highly prevalent chronic disease affecting women in reproductive age. Firstly, the topic of this disease is introduced, the situation and current knowledge in Australia is discussed. Positive changes in a national level are presented. Secondly, early life exposers and maternal behaviour are investigated as possible risk and protective factors. A systematic review of early life factors identified a low birthweight and formula feeding of infants as risk factors for the development of endometriosis. Lastly, the relation of birthweight, weight at childhood and endometriosis was analysed using data of Australian Longitudinal Study on Women's Health. Other risk and protective factors were evaluated and included into the analysis. Logistic regression was used for determination of statistical significance. High weight at 10 years old was found to be a protective factor against endometriosis.
|
|
|
Relation between Economic Situation and Political Stability
JAKLOVÁ, Kamila
The political parties use the economic situation of the state as one of their arguments in their electoral strategy. The aim of the thesis is to depict the impact of the economic situation in selected European countries on voters' decision making. Selected economic indicators, such as gross domestic product, unemployment, inflation, tax on personal income, balance of trade and debt are compared with the impact on voters' decision making in these countries with the application of Logistic Regression. This method determines the degree of impact of individual economic indicators on the political stability of selected countries. The data used in this work are based on the publicly available database of the Organization for Economic Cooperation and Development in years 2007-2017. The result of the analysis determines appropriate economic indicators for political election strategies.
|
|
|
Neural networks and tree-based credit scoring models
Turlík, Tomáš ; Krištoufek, Ladislav (advisor) ; Fanta, Nicolas (referee)
The most basic task in credit scoring is to classify potential borrowers as "good" or "bad" based on the probability that they would default in the case they would be accepted. In this thesis we compare widely used lo- gistic regression, neural networks and tree-based ensemble models. During the construction of neural network models we utilize recent techniques and advances in the field of deep learning, while for the tree-based models we use popular bagging, boosting and random forests ensembling algorithms. Performance of the models is measured by ROC AUC metric, which should provide better information value than average accuracy alone. Our results suggest small or even no difference between models, when in the best case scenario neural networks, boosted ensembles and stacked ensembles result in only approximately 1%−2% larger ROC AUC value than logistic regression. Keywords credit scoring, neural networks, decision tree, bagging, boosting, random forest, ensemble, ROC curve
|
|
|
Landslide susceptibility analysis of Czechia
Racek, Ondřej ; Blahůt, Jan (advisor) ; Klimeš, Jan (referee)
In geosciences modelling is rather quickly developing discipline. Statistical modelling of landslide susceptibility is relatively more traditional approach. Nevertheless, more complicated statistical methods are being developed and applied on larger areas. This development is caused especially by increasing computational capacity and software. This diploma thesis summarises existing statistical landslide susceptibility modelling approaches. In the following part, several landslide susceptibility models were created for the area of Czechia. These models were created using logistic regression, naive Bayes and artificial neural network (ANN). Additionally, two more models were created using expert driven approach. All models were made using thirteen conditioning factors, i.e.elevation, slope, engineering geological regions, climatic areas, mean annual precipitation, topographic wetness index (TWI), aspect, orogenetic class, distance from confirmed fault, distance from watercourse, internal relief, land cover and slope shape. Models driven by statistical approach were created using Orange software. Landslide inventories that were used for construction of all models are based on two databases: "Registr svahových nestabilit" and "Registr sesuvů-Geofond". Using validation by SRC, PRC and ROC curves...
|
|
|
Impacts of 1990's genocide on population of Rwanda and its awareness about family planning
Jelínková, Kamila ; Hulíková Tesárková, Klára (advisor) ; Kurtinová, Olga (referee)
Impacts of 1990's genocide on population of Rwanda and its awareness about family planning Abstract This thesis pursues the demographic development and the population's attitude to family planning in the state of Rwanda. The observed time period includes the second half of the 20th century to the present time. The first aim is the basic description of the demographic development of the state in relation to historical events. It's well known that wars have a negative impact on the population and economics of a state. The civil war broke out which brought big population losses in Rwanda in the 1990's. In this period the demographic revolution proceeded as it had in many other developing countries, which was important for the socioeconomic development of the state, and it was ceased as a consequence of the genocide. The Rwandan government at that time as one of the first governments began with the support of family planning programmes, which could have had a positive impact on the acceleration of the demographic transition. The second aim of the thesis is the determination whether the genocide had an impact on these governmental initiatives and stalled broadening of the awareness of family planning among the population this way, namely by means of media. The last aim of the thesis is the determination whether...
|
|
|
Artificial Intelligence for a Board Game
Tureček, Dominik ; Baskar, Murali Karthick (referee) ; Beneš, Karel (advisor)
This work proposes and implements AI agents for the game Dice Wars. Dice Wars is turn-based, zero-sum game with non-deterministic move results. Several AI agents were created using rule-based approach, expectiminimax algorithm, and logistic regression. To evaluate the performance of proposed agents, an implementation of the game was created. Results of the experiments have shown that it's preferable to play aggressively in two-player games and make more optimal moves in games played with more players. The agent using expectiminimax is able to win more than 60 % of games in 8-player games against random players and wins 21.4 % of games played against a mix of seven other agents created in this work. In two-player setups, the agent using logistic regression with numbers of players' scores and number of dice as features has the best performance and wins 59.4 % of games in average.
|
|
|
Linear Logistic Regression Demo
Bak, Adam ; Kesiraju, Santosh (referee) ; Beneš, Karel (advisor)
This bachelor's thesis deals with the machine learning model logistic regression.The aim is to closely inspect and analyze the workings of this model for classification, in order to be able to provide a learning tool in the form of demonstrative application. All of the mathematical formulae, logistic sigmoid, cross entropy error function and gradient are derived and explained in a concise manner. This thesis also provides some insight into the form of the cross entropy error function in the case of linear logistic regression.
|
|
|
Comparison of statistical methods for the scoring models development
Mrázková, Adéla ; Vitali, Sebastiano (advisor) ; Kopa, Miloš (referee)
The aim of this thesis is to introduce and summarize the process of scoring model development in general and then basic statistical approaches used to resolve this problem, which are in particular logistic regression, neural networks and decision trees (random forests). Application of described methods on a real dataset provided by PROFI CREDIT Czech, a.s. follows, including discussion of some implementation issues and their resolution. Obtained results are discussed and compared.
|
guest ::
