|
|
New Models for Automatic Detection of Performance Degradation
Stupinský, Šimon ; Češka, Milan (referee) ; Rogalewicz, Adam (advisor)
Performance testing is a critical factor in the optimisation of programs during its development, but it is still not so well developed in comparison to functional testing. A framework Perun provides full automation of performance management, thereby contributing to the development of this area. We have introduced three non-parametric approaches to performance data modelling: regressogram, moving average and kernel regression, which were integrated within this framework. We try to achieve appropriate approximations of performance data using these techniques, without the assumption of dependence between two variables, which represents the main advantage in comparison to parametric techniques. Further, we have proposed and implemented two methods for automatic detection of performance changes, which works with all kinds of models within Perun . We have demonstrated our solutions on the real project ( Vim ), and on the set of the experimental cases, in which we compared proposed solutions with existing. We have achieved decreased time processing about two-thirds and an almost triple improvement in the fitness of data modelling with new modelling approaches. The proposed detection methods detected performance degradation of three specific functions in comparison of two different versions of Vim, where was present a known performance issue.
|
|
| |
|
|
Technical Analysis
Krčová, Vendula ; Jedličková, Lucie (referee) ; Novotná, Veronika (advisor)
This diploma thesis is focused on the general characteristics of the technical analysis, definitions of the information sources and technical analyses. Main attention is given to the particular indicators of the technical analysis. Practical part of this thesis includes applying of the defined indicators in the evaluation of the money market situation.
|
|
|
Sensitivity analysis of pressure-time method on measurement uncertainty
Červinková, Kateřina ; Himr, Daniel (referee) ; Habán, Vladimír (advisor)
The pressure-time method is one of two methods of measuring the flow rate on large hydraulic structures applicable to IEC 60041, which is based on the temporal integration of the measured pressure difference and the formation of a water hammer in a closed pipe. The aim of this master thesis is to perform a literature review of this method and to evaluate the flow rate of the measured data. Furthermore, the thesis deals with determination of the sensitivity of the evaluated flow rate to the weights of individual pressure sensors and to numerical modifying of the measured pressures. The first part is made using MS Excel. The flow rate is always evaluates with only one pressure sensor and it is compared with the original flow rate. There is research, how absence of the sensor has an impact on the evaluated flow rate. In the second part of the determination of the sensitivity of the evaluated flow rate, various encroachment (signal smoothing, noise, time delay, frequency band removal) are performed of measured pressure signal in Matlab. Various surrounding influences or sensors failures are simulated.
|
|
|
The Use of Means of Artificial Intelligence for the Decision Making Support on Stock Market
Hrach, Vlastimil ; Budík, Jan (referee) ; Dostál, Petr (advisor)
The diploma thesis deals with artificial intelligence utilization for predictions on stock markets.The prediction is unconventionally based on Bayes' probabilistic model theorem and on its based Naive Bayes classifier. I the practical part algorithm is designed. The algorithm uses recognized relations between identifiers of technical analyze. Concretely exponential running averages at 20 and 50 days had been used. The program output is a graphic forecast of future stock development which is designed on ground of relations classification between the identifiers
|
|
|
Implementation of Statistical Functions Using HLS
Šinaľ, Peter ; Martínek, Tomáš (referee) ; Dvořák, Milan (advisor)
The aim of this thesis was to design and implement selected statistical functions used in technical analysis. I focused on moving averages, Black-Schles model for calculating option prices and Indicator Delta. These functions are through HLS transformed into an appropriate description for programmable FPGA. During the transformation process, emphasis is on low latency and resource consumption. Created solutions demonstrate the potential of HLS. They show complexity of the technical analysis and hardware requirements. Achieved results show high accuracy in the simulations. Deviation from the reference value is approximately 6,615*10e-3. The results also indicate thet that reducing latency does not necessarily cause an increase in the consumption of resources on the chip.
|
|
|
Testing of Indicators for Technical Analysis in Stock Market Trading
Melichar, Josef ; Žák, Jakub (referee) ; Rozman, Jaroslav (advisor)
This thesis deals with testing indicators of technical analysis and their behavior with different kinds of market. In this thesis we tested simple moving averages, exponential moving averages, relative strength index indicator, MACD indicator, and stochastic indicator. At the end we tested a combined indicator, and that was stochastic with MACD. We created an automatic trading system for each of these indicators. From the results we can find out, that by optimizing the technical indicators we can get satisfying results by increasing the profitability. Indicators that seemed non-profitable were in fact profitable after the optimization. However testing parameters of indicators of technical analysis alone is not enough to make a stable profit.
|
|
|
Statistical analysis of data from routing problems
Mazlová, Lenka ; Popela, Pavel (referee) ; Šomplák, Radovan (advisor)
This bachelor thesis focuses on statistical analysis of data retrieved from real waste collection. The main goal is to become familiar with statistical hypothesis testing and model creating typical for mathematical statistics and to determine key parameters that affect routing problems. One of the outputs of this thesis is a Microsoft Excel file that contains statistical tests described in following text.
|
|
|
The effect of air pollution on the incidence of asthma symptoms
Velická, Helena ; Štípek, Stanislav (advisor) ; Bencko, Vladimír (referee) ; Rychlíková, Eva (referee)
The effect of air pollution on the incidence of asthma symptoms MUDr. Helena Velická ABSTRACT The aim of the thesis was to establish the effect of short-term ambient air pollutant concentration changes on asthma exacerbation and symptom variability. The study concerned 147 child patients (age 6 - 18 years) and 304 adult patients (age 19 - 62 years) with confirmed diagnosis of asthma. Their respiratory symptoms and other complaints were recorded in diaries during the heating season (November 2013 - February 2014) in the high- polluted industrial city of Ostrava, Czech Republic. The concentrations of PM10, NO2 and SO2 were measured and provided as smoothed daily maps. GPS coordinates of two addresses of each respondent (the residence and the school/work) were linked with the maps and 24-hour exposure of the respondents to each pollutant was determined, regarding the individuals'daily pattern. The relationships between exposures and health effects were analyzed using Generalized Additive Models (GAM) and expressed as odds ratios per 10 µg/m3 increase in the mean 24-hour exposure at the same day, and also in lag days (1-5), both separately and as moving averages (1-5). Significant associations were found between increase of one- to several days exposure to air pollutants and asthma symptom incidence both in...
|
|
|
Sensitivity analysis of pressure-time method on measurement uncertainty
Červinková, Kateřina ; Himr, Daniel (referee) ; Habán, Vladimír (advisor)
The pressure-time method is one of two methods of measuring the flow rate on large hydraulic structures applicable to IEC 60041, which is based on the temporal integration of the measured pressure difference and the formation of a water hammer in a closed pipe. The aim of this master thesis is to perform a literature review of this method and to evaluate the flow rate of the measured data. Furthermore, the thesis deals with determination of the sensitivity of the evaluated flow rate to the weights of individual pressure sensors and to numerical modifying of the measured pressures. The first part is made using MS Excel. The flow rate is always evaluates with only one pressure sensor and it is compared with the original flow rate. There is research, how absence of the sensor has an impact on the evaluated flow rate. In the second part of the determination of the sensitivity of the evaluated flow rate, various encroachment (signal smoothing, noise, time delay, frequency band removal) are performed of measured pressure signal in Matlab. Various surrounding influences or sensors failures are simulated.
|
guest ::
