|
|
Generalized Wilcoxon Test for Censored Data
Vařejková, Michaela ; Maciak, Matúš (advisor) ; Komárek, Arnošt (referee)
This paper deals with the generalized Wilcoxon test and its use for censored data. The introduction describes standard one-sample and two-samples Wilco- xon tests and their basic properties, censored data and methods of censoring. The main part of the paper is devoted to the introduction of the generalized Wilcoxon test and to its properties. First, a test for singly-censored data is de- scribed; the description of a test for doubly censored data follows. The paper concludes with a simulations part in which statistical properties of the test are demonstrated. The first example compares the generalized test with the stan- dard two-samples Wilcoxon test. The second example shows how the censoring rate affects the power and significance level of the generalized test. 1
|
|
|
Estimation of probability distribution for censored data
Teichmannová, Zuzana ; Lachout, Petr (advisor) ; Antoch, Jaromír (referee)
In this thesis, we look into estimation of probability distribution for censored data. These data are not complete, because for some reason it was impossible to observe them all. We use the Kaplan-Meier estimator and study some of its properties. We also use the Nelson-Aalen estimator. In the end we make a compa- rison of these estimators with a naive estimator, which omits the censored data. The comparison is illustrated on two numerical examples where we can see the main differences in the accuracy of the estimators. We will see that it is better to include the censored data to our estimations. 1
|
|
|
On quantile optimization problem with censored data
Volf, Petr
In the framework of stochastic optimization the criterion based on selected quantiles is considered. Further, stochastic characteristics are estimated from censored data. Therefore, certain theoretical results concerning estimators of distribution function and quantiles under censoring are recalled and utilized to prove consistency of solution based on estimates. Behavior of solutions for finite data sizes is studied with the aid of randomly generated example.
|
|
|
On problem of optimization under incomplete information
Volf, Petr
The paper studies consequences of incomplete information to uncertainty of results of stochastic optimization. Stochastic characteristics of optimized system are evaluated from observed data, moreover, the data may be incomplete. Namely, we consider the random censoring of observations frequently encountered in time-to-event (of lifetime) studies. The analysis of uncertainty will be based both on theoretical properties of estimated stochastic characteristics and on simulated examples.
|
|
|
Length of doctoral studies at the Faculty of Informatics and Statistics
Hybšová, Aneta ; Malá, Ivana (advisor) ; Čabla, Adam (referee)
This thesis describes the survival analysis, exactly Kaplan-Meier estimate. A main part of the thesis deals with the problem of censored data, which is typical for survival analysis. The empirical part describes lenght of PhD studies at the Faculty of Informatics and Statistics and their "survival" in studies by Kaplan-Meier curves. First are analyzed uncensored data and then the whole data set (censored and uncensored data).
|
|
|
Estimates in Survival Analysis
Čabla, Adam ; Malá, Ivana (advisor) ; Tomášek, Ladislav (referee)
This thesis introduces methods used in time-to-date analysis. It is written generally and so usable in dealing with any example. The thesis deals with problem of censoring, which means, that some observations occurred after the following, which is typical for the lifetime analysis. Methods mentioned in the thesis are nonparametric and parametric estimates of the survival function and their characteristics, and regression models, concretely Cox model and accelerated failure time model, which examine effect of the covariates on survival function. In the thesis is beside survival function presented hazard function, which express intensity of the analyzed event and cumulative hazard function, which is created as the name suggests by cumulative summation of the hazard function. Estimates of these functions are obtainable from survival function and for parametric estimate often exists formula resulting from parameters of used distribution. Empirical part of the thesis introduces influence of several different types and degrees of censoring on parametric and nonparametric estimates of the survival function, mean and median. The other empirical example is the usage of regression analysis on the data from the lungs cancer research made by Mayo Clinic.
|
guest ::
