National Repository of Grey Literature 43 records found  beginprevious30 - 39next  jump to record: Search took 0.01 seconds. 
Nonlinear regression in R programming langure
Dolák, Martin ; Malá, Ivana (advisor) ; Bašta, Milan (referee)
This thesis deals with solutions of nonlinear regression problems using R programming language. The introductory theoretical part is devoted to familiarization with the principles of solving nonlinear regression models and of their applications in the program R. In both, theoretical and practical part, the most famous and used differentiator algorithms are presented, particularly the Gauss-Newton's and of the steepest descent method, for estimating the parameters of nonlinear regression. Further, in the practical part, there are some demo solutions of particular tasks using nonlinear regression methods. Overall, a large number of graphs processed by the author is used in this thesis for better comprehension.
Utilizing Bootstrap and Cross-validation for prediction error estimation in regression models
Lepša, Ondřej ; Bašta, Milan (advisor) ; Malá, Ivana (referee)
Finding a well-predicting model is one of the main goals of regression analysis. However, to evaluate a model's prediction abilities, it is a normal practice to use criteria which either do not serve this purpose, or criteria of insufficient reliability. As an alternative, there are relatively new methods which use repeated simulations for estimating an appropriate loss function -- prediction error. Cross-validation and bootstrap belong to this category. This thesis describes how to utilize these methods in order to select a regression model that best predicts new values of the response variable.
Neural Networks in R
Arzumanov, Eduard ; Bašta, Milan (advisor) ; Žižka, David (referee)
The aim of this work was to present the issue of neural network, which is still, despite the fact it exist and has been applied for several years, remains quite unknown for a considerably big part of public and academical environment. The aim of the practical part was to verify via practical application if neural network are truly a better instrument of statistical analysis, than the commonly used ones, especially when the goal is to analyze and describe complex processes and relationships between them. Further aim of the work was to investigate and describe the relationships between the development of trading volumes of Apple shares and the shares of competitive companies regarding the market of smart phones such as Google, HTC, Nokia, Samsung using neural network models. The attainment of these goals was realized through a rather extensive description of neural networks theory as well as the presentation of valuable theoretical tools for avoiding the frequent barriers occurring during the practical implementation. This practical application was realized via software called R, which has widely spread lately due to its availability and a vast range of flexibility, which is provided to users. The value of this work is familiarization and the creation of an integrated knowledge within readers about the issue of neural networks and the deliverance of a proof, that neural networks are indeed a better tool compared to the commonly used ones (ARMA models, linear regression). The author of the work gained a lot of useful knowledge about neural networks, learned how to use them in practice especially in the environment of R software, by which he shifted his proficiency with the current software to a whole new level.
The methods for detection of the outliers and influential points based on method of least squares in linear regression analysis. The qualitative comparison with the detection methods based on robust regression.
Potůčková, Lenka ; Bašta, Milan (advisor) ; Blatná, Dagmar (referee)
This Thesis deals with the methods for detection of the outliers and influential points based on method of least squares. The first part of the thesis summarizes the teoretical findings of the method of least squares and both methods for detection of the outliers and influential points based on the method of least squares and also based on robust regression. The practical part of this thesis deals with the application of classic methods for detection of the outliers and influential points on three types of datasets (artifical data, data from specialized literature and real data). The results of the application are subject to qualitative comparisson with the results produced by the methods for detection of the outliers and influentials point based on the robust regression.
Potential output. Econometric application for Czech Republic.
Kyncl, Jan ; Pánková, Václava (advisor) ; Bašta, Milan (referee)
I summarize different methods of potential output and output gap estimation including advantages and disadvantages in this thesis. I also applied two published models on real data for Czech Republic. Concerned models are Hodrick-Prescott filter and so called Production Approach. Both approaches are simultaneously used by ČNB. This thesis offers comparison between HP filter and production approach and comparison of Czech, Austrian and common EU-15 potential output and output gap. Potential output of Austria and EU-15 was obtained from OECD database. Comparison result refers to very similar progress of estimate obtained by univariate and multivariate method. It also shows different trend behavior of domestic economy against more developed EU countries, which is starting to be similar at the end of observed period.
Seasonality on selected demographic time series over the last 10 years
Bezchlebová, Daniela ; Šimpach, Ondřej (advisor) ; Bašta, Milan (referee)
Nowadays there is a noticeable trend towards increasing number of children being born out of wedlock -- that is, not in marriage. But the term "marriage" has gained a completely different meaning than in the past. Unmarried cohabitation has become a common alternative to marriage. This is illustrated by the fact that until the end of the 1960s, the proportion of births out of wedlock is very low. The aim of this thesis is to analyse the monthly time series of the number of in-wedlock and out-of-wedlock live births in the Czech Republic over the last ten years, based on selected socio-demographic and technical characteristics, and to prove or reject the statistical significance of the occurrence of seasonal element and further quantify and simulate this element.
Currency Trading Strategies
Krpálek, Jan ; Bašta, Milan (advisor) ; Žváčková, Lenka (referee)
My bachelor thesis is concerned with algorithmic trading on foreign exchange markets. Motivation is to create practical work which will cover basics for my ongoing work. From my perspective it is very important to describe assumptions which are necessary in order to properly understand functionality of automated trading. The initial theoretical part includes essential description of methods and tools for Technical Analysis and Money Management which are used afterwards. Further my aim is to show optimization process that is used by professional traders to achieve better and consistent trading results. Finally all calculations will be processed by the TradeStation trading platform and written in EasyLanguage. The end of my thesis includes complete algorithm, ready for immediate use.
Analysis of climatological time series for selected places in Europe in the years 1961-2010
Míka, Tomáš ; Helman, Karel (advisor) ; Bašta, Milan (referee)
This bachelor thesis preoccupied with the analysis climatological time series mesured at seven selected meterological stations located throughout the European continent. The underlying data for this analysis were obtained from publicly available European database of ECA&D. The analysis is focused on calculation of the basic charakteristic and decomposition of time series. The aim of this bachelor thesis is to examine development of the time series of average monthly temperatures and precipation in the period 1961 - 2010 for individual weather stations and these data analyzed to compare with each other and explore similiraties and differences in temperature and precipation from individual station, not only in term of time but also terms of space. This thesis is divided into two parts. The first part is theoretical and generally discribes statistical methods. The second key part (practical) is devoted data processing and analysis of the time series.
Wavelet Transform and its Application in the Analysis of Economic and Financial Time Series
Bašta, Milan ; Arlt, Josef (advisor) ; Málek, Jiří (referee) ; Mareš, Milan (referee)
The thesis deals with a brief compilation of the theory of Fourier transform, linear filtration and a triad of wavelet transforms -- the maximal overlap discrete wavelet transform (MODWT), the discrete wavelet transform (DWT) and the continuous wavelet transform (CWT). These transforms are among others applied to the analysis of the time-varying character of variability in the time series, to the detection of events of significant changes of variability, to the removal of noise in the time series (denoising) and to the time-scale analysis of the relationship of two time series. The analyzed time series used in this thesis are the logarithm of the Garman-Klass estimate of the historical volatility, the time series of stock returns and the logarithm of the monthly inflation rate. In some cases artificial time series are analyzed. The procedures and methods introduced in the thesis might be well implemented in the analysis of other economic and financial time series. The contribution of the thesis is a brief and easy-to-use compilation of the wavelet theory and the application of the wavelet transform to such financial and economic time series, where such an analysis tool has never been applied before. New insights into the properties of time series are thus obtained, insights, which might be hardly recovered by traditional means and methods.
The study of the relationship between average realized stock returns and the risk of stock investment
Řípa, Daniel ; Bašta, Milan (advisor) ; Cícha, Martin (referee)
Bachelor thesis deals with the topic of average return rates of stock investments and assessment of their risk. The aim of this work is a comparison of two alternative approaches of risk measurement. Twenty years long time series of 30 stocks' returns from 1991 to 2010 are first used to analyze a relationship between average realized returns and standard deviations of the returns. Variety of computational algorithms is used in attempt to generalize this relationship. Analysis of the full length time series does not seem to discover a significant mutual relationship in the analyzed dataset. However, by analogically employing the algorithm of the CAPM analysis a significant and positive linear relationship between standard deviations and realized returns was found. Furthermore, two-step regression algorithm introduced by Fama & MacBeth is used to test the validity of CAPM model. This method led to the conclusion that CAPM cannot be rejected within the analyzed dataset. Moreover, strong positive linear relationship was found between the estimates of standard deviation and beta coefficients, which may be explained by the lack of variability in correlation between individual assets' and market portfolio's returns.

National Repository of Grey Literature : 43 records found   beginprevious30 - 39next  jump to record:
See also: similar author names
1 Bašta, Martin
2 Bašta, Miroslav
Interested in being notified about new results for this query?
Subscribe to the RSS feed.