|
|
Analysis of Selected Climatological Time Series
Kraitz, Petr ; Helman, Karel (advisor) ; Šimpach, Ondřej (referee)
Dissertation is focused on analysis of time series of monthly mean temperatures of seven meteorological stations between years 1960 to 2015. With the use of selected statistical methods time series are modelled in moving ten year long intervals and a forecast of elevenths year is constructed. The goal of this dissertation is to compare the success of different methods/models predictions of values for one year period, based on the data from last 10 years.
|
|
|
Analysis of Wages and Salaries in the Pardubice Region
Hanuš, Pavel ; Löster, Tomáš (advisor) ; Šimpach, Ondřej (referee)
This bachelor thesis deals with the study of changes in income of the economically active population in the given region. The main aim of the thesis is to analyze the wage and salary sphere of the population in the Pardubice region and to compare the size of wages and salaries by category. These include age groups, gender, highest education and type of profession. The partial aim is to predict the evolution of hourly earnings using appropriately selected statistical methods. The main benefit of this thesis is to provide a comprehensive overview of the size of the personal income in the Pardubice region for 2015 and 2016.
|
|
|
Analysis of salaries and wages in the capital city of Prague
Jermář, Martin ; Löster, Tomáš (advisor) ; Šimpach, Ondřej (referee)
The aim of the bachelor thesis is to analyse the development of wages and salaries in Prague, the capital city of the Czech Republic by using statistical methods. The thesis includes a general description of terms of salary and wage sphere, time series and selected applied statistical methods. In practical part the analysis of development of wages and salaries in Prague will be performed using the rates of dynamics. Further, time series predictions will be created using adaptive methods. The analysis should familiarize the reader with wage and salary developments for the period from 2011 to 2016 and outline the predictions of possible values in 2017. The results of the thesis show a higher increase in average and median monthly earnings in a wage sphere, increasing disproportions in monthly earning between men and women and bigger disproportions in monthly earnings in the wage sphere. Based on predictions performed the growth of wages and salaries is expected in 2017.
|
|
|
An influence of offline advertising on website traffic
Solnický, Vojtěch ; Löster, Tomáš (advisor) ; Šimpach, Ondřej (referee)
The time series of sessions on web site is investigated in the paper with using box and line plots. On organic traffic, which is part of the main time serie of sessions, is aplicated F-test of seasinality which proves presence of seasonality. The seasonality is removed from time searie by X-13 ARIMA method. The influence of offline advertising on organic traffic of the website is investigated in the second part of this work by using ADL model, but firstly all time series are tested on stacionarity by Dickey-Fuler test which proves, that they are all stacionary. The final model is tested on heteroscedasticity by ARCH test with no prove of it, on autocorelation by BG test also with no prove of it and by Jarque-Bera test on normal distribution of residuals, which proves that they follow normal distribution. The result from test says, that the time searie of organic traffic of the website is influenced by television advertising and by lagged organic trafic itslef with lag of one month.
|
|
|
Analysis of time series of spa guests in the Czech Republic
Žák, Václav ; Matějka, Martin (advisor) ; Šimpach, Ondřej (referee)
The bachelor thesis is focused on analysis of quarterly time series of guests in spa resorts in Czech republic in the period from 2000 to 2016. The data were obtained from Czech statistical office. The aim of this thesis is to analyse the time series and to create predictions of possible future progress of time series. Deterministic and stochastic methods are used to analyse the time series. All models are created in the statistical environment R.
|
|
|
Comparing the effectiveness of selected methods of cancer treatment. Prostate cancer, breast cancer and lung cancer via survival analysis
Šimonková, Karolína ; Šimpach, Ondřej (advisor) ; Pechholdová, Markéta (referee)
This diploma thesis deals with various ways of treatment of selected oncological diseases and the effectiveness of treatment methods and evaluation of the influence of various factors influencing the survival of patients. The activity of individual healing processes is evaluated by survival analysis. The subjects of the study are patients with breast, lung and prostate cancer. The survival analysis considers the sex of the patient, the age and stage of his illness, and other factors to avoid distorted results. The aim of the work is to find out the effects of selected therapeutic procedures on patients' health and to identify factors that have a significant impact on the survival of patients. The data for the diploma thesis was provided by the Institute of Health Information and Statistics of the Czech Republic, the Statistical Office, the National Cancer Register (NOR), the US SEER database and the German Breast Cancer Study.
|
|
|
Changes in development of fertility and birth rates depending on the economic conditions in the Czech Republic
Sudová, Petra ; Arltová, Markéta (advisor) ; Šimpach, Ondřej (referee)
This diploma thesis deals with the analysis of the birth rate and fertility rate in dependence on the economic conditions of the Czech Republic. The aim of this thesis is to analyze relations between selected socio-economic indicators and total fertility rate on the basis of available data and to evaluate changes related to birth and fertility in the Czech Republic, which occurred in the period 1993-2015. The thesis is divided into two main parts - theoretical and analytical. In the theoretical part are described the basic methods of calculating the characteristics for the analysis of the level of birthrate and fertility, as well as the development of selected socio-economic indicators. An important part of the first part of the diploma thesis is specification of used methods within time series. The second part is practically focused on cointegration analysis and subsequent assembly of single-row models from which error correction models were obtained by transformation. These can be used to describe short and long term relationships between time series. Explained variables are aggregate fertility in all assembled models, the explanatory variables are GDP, average gross monthly wage, final consumption expenditure for households, child allowances, parental allowance and household loans per capita.
|
|
|
The consumer loans analysis using statistical methods
Božíková, Barbora ; Löster, Tomáš (advisor) ; Šimpach, Ondřej (referee)
Consumer loans are part of loan products provided by bank institutions. This diploma thesis is focused on possibility of identifying risk clients with the consumer loans, using available data set. In the first part of the work was briefly mentioned the credit process and also theoretical basis of statistic methods used in empirical part of the work. In the second part were investigated dependencies, and was described the clients structure. Then the discriminant analysis was applied, with the aim to identify the sorting criteria, which could recognize the risk and unproblematic clients. Subsequently the results of the analysis were evaluated and described the identified connections.
|
|
|
Comparison of Differences in Seasonality of Demographic Time Series of The Selected Countries in EU
Morávek, David ; Šimpach, Ondřej (advisor) ; Miskolczi, Martina (referee)
In the theoretical level this thesis describes some of the statistical methods using for time series analysis with seasonal component and some of the statistical methods using only for analysis of seasonal component. In the analytical level, these statistical methods are applied to demographic time series. The seasonal component is analysed in more detail, and the progress and type of the seasonal component is compared between several european country. This thesis brings and shows some of the interesting trends, which are appeared in analysed demographic time series. Last but not least, this thesis also provides the preview on the application of some not usually using statistical methods.
|
|
|
The analysis of marriage and divorce rates in the Czech Republic in confrontation with Greece
Davidová, Monika ; Šimpach, Ondřej (advisor) ; Antovová, Michaela (referee)
Bachelor thesis follows up demographic development of marriage and divorce rates in Czech Republic and Greece during 1992-2012. The main diference is the religion. Greece is high-believer country. In the first chapter are defined basic terms and formulas which are used in analysis of results. The second chapter is about methodics. This part talks about the decomposition of time series and trend functions. In the third chapter are evaluated models. Analytical chapter shows reasons why values have this results. Next point is the comparing of crude rates, total divorce rate or proportions of children born outside marriage. Last subchapter is the forecasts of future development based on modeling trends and exponential smoothing. Bachelor thesis is creating summary view of development in marriage and divorce rates in Czech Republic and Greece.
|
guest ::
