|
|
Statistical Methods for Regression Models With Missing Data
Nekvinda, Matěj ; Kulich, Michal (advisor) ; Omelka, Marek (referee)
The aim of this thesis is to describe and further develop estimation strategies for data obtained by stratified sampling. Estimation of the mean and linear regression model are discussed. The possible inclusion of auxiliary variables in the estimation is exam- ined. The auxiliary variables can be transformed rather than used in their original form. A transformation minimizing the asymptotic variance of the resulting estimator is pro- vided. The estimator using an approach from this thesis is compared to the doubly robust estimator and shown to be asymptotically equivalent.
|
|
|
Post-selection Inference: Lasso & Group Lasso
Bouř, Vojtěch ; Maciak, Matúš (advisor) ; Kulich, Michal (referee)
The lasso is a popular tool that can be used for variable selection and esti- mation, however, classical statistical inference cannot be applied for its estimates. In this thesis the classical and the group lasso is described together with effici- ent algorithms for the solution. The key part is dedicated to the post-selection inference for the lasso estimates where we explain why the classical inference is not suitable. Three post-selection tests for the lasso are described and one test is proposed also for the group lasso. The tests are compared in simulations where finite sample properties are examined. The tests are further applied on a practical example. 1
|
|
|
Confidence Intervals for Quantiles
Horejšová, Markéta ; Kulich, Michal (advisor) ; Hlávka, Zdeněk (referee)
In this thesis, various construction methods for simultaneous confidence intervals for quantiles are explained. Among nonparametric approaches, a special emphasis is dedicated to a recent method based on a multinomial distribution for calculating the overall confidence level of confidence intervals for all quantiles of interest using an efficient recursive algorithm, which is also described. Furthermore, a method based on Kolmogorov-Smirnov statistic or an asymptotic method using empirical distribution function and order statistics for quantile estimate are presented. A special parametric method for several quantiles of a normally distributed population is introduced along with a few of its modifications. Subsequently, a simulation is run to test the real coverage of the described theoretical methods. Powered by TCPDF (www.tcpdf.org)
|
|
|
Statistical analysis of datasets with missing observations
Janoušková, Kateřina ; Omelka, Marek (advisor) ; Kulich, Michal (referee)
Mechanisms of missing data and methods are described in this thesis. Three mechanisms are considered - MCAR, MAR, MNAR. Two simple methods using deletion of incomplete records are shown and their properties and shortcomings are demonstrated. Secondly, the principle of simple imputations is explained. EM algorithm which uses the classical statistics and the algorithm of data augmentation which uses Bayesian framework are derived and compared. The last method included in the thesis is the multiple imputation. The described methods are compared on real data set, first on continuous variables and then on a contingency table. 1
|
|
|
Models for zero-inflated data
Matula, Dominik ; Kulich, Michal (advisor) ; Hlubinka, Daniel (referee)
The aim of this thesis is to provide a comprehensive overview of the main approaches to modeling data loaded with redundant zeros. There are three main subclasses of zero modified models (ZMM) described here - zero inflated models (the main focus lies on models of this subclass), zero truncated models and hurdle models. Models of each subclass are defined and then a construction of maximum likelihood estimates of regression coefficients is described. ZMM models are mostly based on Poisson or negative binomial type 2 distribution (NB2). In this work, author has extended the theory to ZIM models generally based on any discrete distributions of exponential type. There is described a construction of MLE of regression coefficients of theese models, too. Just few of present works are interested in ZIM models based on negative binomial type 1 distribution (NB1). This distribution is not of exponential type therefore a common method of MLE construction in ZIM models cannot be used here. In this work provides modification of this method using quasi-likelihood method. There are two simulation studies concluding the work. 1
|
|
|
Statistical analysis of datasets with missing observations
Janoušková, Kateřina ; Omelka, Marek (advisor) ; Kulich, Michal (referee)
Mechanisms of missing data and methods of their treatment are de- scribed in this thesis. Three mechanisms are considered - MCAR, MAR, MNAR. Two simple methods using deletion of incomplete records are introduced and their properties and shortcomings are described. Further, the principle of simple imputations is explained. EM algorithm which uses the classical statistics and the algorithm of data augmentation based on Bayesian framework are derived and compared. The last method included in the thesis is the multiple imputation. The described methods are applied on real data set, first on continuous variables and then on a two dimensional contingency table. 1
|
|
|
GOF tests for gamma distribution
Klička, Petr ; Hlávka, Zdeněk (advisor) ; Kulich, Michal (referee)
The Bachelor thesis deals with the goodness of fit test for the Gamma distribution. Initially, we show several ways how to estimate the parameters of the Gamma distribution - firstly, the maximum likelihood estimator is presented, followed by estimator gained by the method of moments and fi- nally, we introduce the new estimator based on the sample covariance. The last estimator is used for constructing the goodness of fit test for the Gamma distribution. We define the test statistics V ∗ n to this test and its asymptotic normality is derived under the assumption of the null hypothesis. At the end of the thesis the simulations are realized to obtain the empirical size of the test for various values of parameter a and parameter b which equals one. 1
|
|
|
Confidence Intervals for Binomial Parameters
Rusá, Pavla ; Kulich, Michal (advisor) ; Maciak, Matúš (referee)
The Bachelor thesis deals with the construction of confidence intervals for the parameter of the Binomial distribution. In the first part of the thesis we deal with the relationship between hypotheses testing and confidence in- tervals. The methods mentioned in this thesis are based on this relationship. The next part is devoted to the Clopper-Pearson method and its possible improvements which were made by Sterne, Crow, Blyth and Still. The gra- phical approach of Schilling and Doi is also worth noticing. Afterwards we present the methods based on the Normal approximation, particularly the Wilson method and the Wald method. Finally, all the methods mentioned in this thesis are compared in terms of coverage probability. 1
|
|
|
Variable selection based on penalized likelihood
Chlubnová, Tereza ; Kulich, Michal (advisor) ; Maciak, Matúš (referee)
Selection of variables and estimation of regression coefficients in datasets with the number of variables exceeding the number of observations consti- tutes an often discussed topic in modern statistics. Today the maximum penalized likelihood method with an appropriately selected function of the parameter as the penalty is used for solving this problem. The penalty should evaluate the benefit of the variable and possibly mitigate or nullify the re- spective regression coefficient. The SCAD and LASSO penalty functions are popular for their ability to choose appropriate regressors and at the same time estimate the parameters in a model. This thesis presents an overview of up to date results in the area of characteristics of estimates obtained by using these two methods for both small number of regressors and multidimensional datasets in a normal linear model. Due to the fact that the amount of pe- nalty and therefore also the choice of the model is heavily influenced by the tuning parameter, this thesis further discusses its selection. The behavior of the LASSO and SCAD penalty functions for different values and possibili- ties for selection of the tuning parameter is tested with various numbers of regressors on simulated datasets.
|
|
|
Models with categorical response
Faltýnková, Anežka ; Kalina, Jan (advisor) ; Kulich, Michal (referee)
This thesis concentrates on regression models with a categorical response. It focuses on the model of logistic regression with binary response and its generalization in which two models are distinguished: multinomial regression with nominal response and multinomial regression with ordinal response. For all three models separately, the Wald test and the likelihood ratio test are derived. These theoretical derivations are then used to calculate the test statistics for specific examples in statistical software R. The theory described in the thesis is illustrated by examples with small and large number of explanatory variables.
|
guest ::
RSS feed.