|
|
Modern statistical approach in evaluating the compressive strength of concrete in structures using the rebound hammer method
Janka, Marek ; Kocáb, Dalibor (referee) ; Misák, Petr (advisor)
This diploma thesis examines various linear regression methods and their use to establish regression relationships between the compressive strength of concrete determined by the indirect method and by the crushing of the specimens in the press. It deals mainly with the uncertainty of values measured by the indirect method, which is neglected by the usually used ordinary least squares regression method. It also deals with the weighted least squares method, suitable for so-called heteroskedastic data. It compares different regression methods on several sets of previously measured data. The final part of the work examines the effect of removing too influential points identified by Cook's distance, which may skew the regression results.
|
|
| |
|
|
Robust Student estimator
Hlavinka, Radek ; Friml, Dominik (referee) ; Dokoupil, Jakub (advisor)
This Master's thesis deals with Bayesian approach to robust parameter estimation for ARX models. Robustness is achieved by assuming the measurement noise to be generated by Student-t distribution. The asumption of Student-t noise renders the model's posterior intractable and requires utilization of approximation techniques. This thesis considers algorithms using Gibbs sampler and Variational approximation and compares them with Ordinary Least Squares. The algorithms are compared based on their Maximum Likelihood estimation. It is shown that approaches assuming the Student-t noise perform better in simulation. The results from data acquired from physical system are however similar for all algorithms considered.
|
|
|
The impact of COVID-19 on election outcomes: The case of the Czech Republic
Válek, Vojtěch ; Pertold-Gebicka, Barbara (advisor) ; Mošovský, Jan (referee)
This thesis examines the relationship between the COVID-19 pandemic and the election results in the 2021 parliamentary elections in the Czech Republic. Voter turnout together with the results of several political parties is analyzed at the district level. First, the election results of 2021 are examined separately using OLS regression, heteroskedasticity is not detected in the model. A panel dataset is then created, which also includes election results from 2013 and 2017. Based on the results of the Hausman test, the fxed efects method is used to analyze the panel data, and the frst-diferencing method is used as well for comparison. Pairs of years are then analyzed with the use of frst-diferencing, when the results are frst estimated without 2013, and then without 2017. The results of the analysis mainly show a signifcant relationship between the number of infected people and the results of individual political parties. While for political party ANO this relationship is negative, it is rather positive for the other political parties. The second dependent variable of interest is the number of deaths related to COVID-19. In this case, the results of the analysis do not indicate a signifcant connection with the election results. 1
|
|
|
Estimation and goodness-of-fit criteria in logistic regression model
Ondrušková, Markéta ; Hanzák, Tomáš (advisor) ; Zvára, Karel (referee)
In this bachelor thesis we describe binary logistic regression model and estimation of model's parameters by maximum likelihood method. Then we propose algorithm for the least squares method. In the goodness-of-fit criteria part we define Lorenz curve, Gini coefficient, C-statistics, Kolmogorov-Smirnov statistics and coefficient of determination R2 . We derive their relation to different sample coefficients of correlation. We derive typical relation between Gini coeffi- cient, Kolmogorov-Smirnov statistics and newly also coefficient of determination R2 via model of normally distributed score of bad and good clients. These derived teoretical results are verified on three real data sets. Keywords: Binary logistic regression, maximum likelihood, ordinary least squa- res, Gini coefficient, coefficient of determination. 1
|
|
|
Least Squares Alternatives
Gerthofer, Michal ; Pešta, Michal (advisor) ; Kulich, Michal (referee)
In the present thesis we deal with the linear regression models based on least squares. These methods are discussed in two groups. The first one focuses on three primary aproaches devided by occurrence of errors in variables. The traditional approach penalizes only the misfit in the de- pendent variable part and is called the ordinary least squares (OLS). An opposite case to the OLS is represented by the data least squares (DLS), which allow corrections only in the explanatory variables. Consecutively, we concentrate ourselves on the total least squares approach (TLS) mi- nimizing the squares of errors in the values of both dependent and independent variables. Finally, we give attention to next group of methods whit high breakdown point, which deal with signifi- cance of the individual observations (least weighted squares) and elimination of outlying obser- vations (least trimmed squares). The main purpose of this work is to describe and compare these models, their assumptions, characteristics, properties of estimates and show them on real data. 1
|
|
|
Modern statistical approach in evaluating the compressive strength of concrete in structures using the rebound hammer method
Janka, Marek ; Kocáb, Dalibor (referee) ; Misák, Petr (advisor)
This diploma thesis examines various linear regression methods and their use to establish regression relationships between the compressive strength of concrete determined by the indirect method and by the crushing of the specimens in the press. It deals mainly with the uncertainty of values measured by the indirect method, which is neglected by the usually used ordinary least squares regression method. It also deals with the weighted least squares method, suitable for so-called heteroskedastic data. It compares different regression methods on several sets of previously measured data. The final part of the work examines the effect of removing too influential points identified by Cook's distance, which may skew the regression results.
|
|
|
Analysis of Short-term Rental Market in Prague
Štollová, Karolína ; Pertold-Gebicka, Barbara (advisor) ; Pavlíček, Jaroslav (referee)
This paper investigates determinants of daily rate of Airbnb listings in Prague, Czech Republic. Sample of 13 500 properties was examined to identify a relationship between property attributes and rental price using the ordinary least squares estimation method. The study provides an empirical evidence that twenty-five independent variables describing space, reputational, location, commerciality attributes or management policies significantly impact the average daily rate. According to the analysis the most relevant rental price determinants are property location, its size in terms of number of bedrooms, bathrooms and capacity. Author demonstrated in two robustness checks that the results are stable. The study is conducive to better understanding of the Prague Airbnb market. Insights from the analysis could help hosts in developing a suitable pricing strategy as well as Airbnb or similar platforms in designing a pricing tool to increase hosts efficiency.
|
|
|
School bullying and its effect on children's academic performance in the Czech Republic
Kalendová, Tereza ; Chytilová, Julie (advisor) ; Polák, Petr (referee)
A safe school environment is a crucial aspect for sustainable educational quality as well as for promoting students' academic performance. One of the factors damagingly affecting school environments is school bullying. Researchers investigating bullying problematic have revealed many negative consequences of bullying on students' general well-being, including its adverse effect on school performance. The thesis aims to investigate the impact of school bullying on student's academic achievement in the Czech Republic, using mathematics test score results of fourth grade students from TIMSS 2015. In the first stage, we use ordinary least squares (OLS) regression controlling for a number of student's background variables. As the estimates obtained from this method might be biased due to endogeneity issues, we apply non-parametric technique propensity score matching (PSM) in the second stage. Our findings suggest that school bullying has a significant negative effect on students' mathematics test scores. However, the effect loses its significance as the frequency of bullying decreases. The PSM technique yield similar estimates as the OLS. The results indicate that Czech educators and policymakers should take into account bullying as an important determinant of pupils performance and enhance...
|
|
|
Empirical Analysis of Prague Flat Market
Sklenářová, Tereza ; Křehlík, Tomáš (advisor) ; Vozková, Karolína (referee)
The purpose of this work is to model the prices of real estate, concretely of Prague flats, which belong to the most important economic indicators. In the theoretical part, the main housing market participants are defined, special features of housing markets are described and most frequently used valuation methods are discussed. Most attention is focused on so called hedonic pricing model, which is applied as a base for the pricing equation in the econometric part. This is carried on various subsets of public available data regarding the characteristics of Prague flats, using ordinary least squares as well as weighted least squares. Several hypotheses about the relationship between the price and the explanatory variables are tested before creating the final model. The results are commented and compared with literature concerned with the same topic in other locations.
|
guest ::
