Academic literature on the topic 'Náhodná chyba'

V článku se věnujeme srovnání klasické a bayesovské exploratorní faktorové analýzy. Použili jsme náhodně generovaná testovací data s různými velikostmi vzorku, strukturou faktorových nábojů, škálami odpovědí a velikostmi náhodné chyby. Provedli jsme oba druhy exploratorní faktorové analýzy a na základě srovnání výsledků a celkové zkušenosti z provedených analýz můžeme uzavřít, že u klasické a bayesovské exploratorní faktorové analýzy není rozdíl v přesnosti odhadu faktorových nábojů. Bayesovský přístup poskytuje více informací, které jsou intuitivněji interpretovatelné, ale na druhou stranu námi použitý model má i nevýhody, jako je nutnost ruční úpravy zdrojových kódů a časová náročnost výpočtů. Bayesovský přístup je tak podle nás vhodnější spíše pro konfirmatorní faktorovou analýzu.

This thesis is focused on problem data reconciliation of measurements. The objective of this thesis was reconciled measured value from electric drum dryer to suit exactly to the mathematical model of drying. For solution was used nonlinear data reconciliation with constrained nonlinear optimization. The entire calculation is processed in programme MATLAB and outputs are graphs of reconciled values of measurement on dryer such as inlet and outlet temperature and humidity, differential pressure of exhaust moisture air, weight of laundry, atmospheric pressure and electric supply. Achieved solution can by characterized by an amount of evaporated water. Weight of wet and dry laundry are 27,7 kg a 17,7 kg. The calculated amount of evaporated water from measurements was almost 18,8 kg. With reconciled measurements it was 9,7 kg. Goals of the thesis were found more realistic values.

The diploma thesis consists of a new design of a controlling procedure for a plastic component of a tank. Its theoretical part deals with the potential sources of errors and uncertainties arising from the measuring itself. It also deals with the characteristics of available CMM devices and includes an analysis of the methods assessing the acceptability of the measurement plan. The practical part analyses systematically the current state and proposes a new solution using a program created using a coordinate machine. A comparison of both variants is presented as an output of the thesis.

1 ABSTRACT Background The annual drug overdose rates have been increasing exponentially since the 90's worldwide. Toxicological information centre (TIC) represents a valuable source of information for evaluating the trends in the drug poisonings in Czech Republic. Aim of the study The purpose was to analyze the number and trends in the calls concerning poisonings due to central nervous system (CNS) affecting drugs, identify the reasons of medication errors caused by laymen, frequency and consequences of these errors across all age groups and also to analyze the numbers, causes, symptoms and severity of the paracetamol intoxications. Methods During the reference period the data from the enquiries were extracted from the TIC electronic database, discharge reports from the hospital were studied or phone call follow-ups with the patients were carried out to be able to evaluate the outcome of the poisonings. Results In the years 1997-2002 the number of calls caused by poisoning with tricyclic antidepressants and barbiturates decreased (by 366.7 % and 340,0 %, respectively) whereas the calls due to selective serotonin reuptake inhibitors and benzodiazepines overdose increased (by 1347.4 % and 359.8 %). The 0-5 years old children are at the highest risk of experiencing medication errors or accidental poisonings...

This work investigates mainly panel data models in which cross-sections can be considered independent. In the first part, we summarize results in the field of pool models and one-way error component models with fixed and random effects. We focus especially on the ways of estimating unknown parameters and on effects significance tests. We also briefly describe two-way error component model issues. In the second part, estimators of first order autoregressive panel data model parameters are derived, for both fixed and random parameters case. The work proves unbiasedness, consistency and asymptotic normality of selected estimators. Using these features, hypothesis tests about corresponding parameters are derived. Application of models is illustrated using real data and simulated data examples. Powered by TCPDF (www.tcpdf.org)