• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 2
  • 1
  • Tagged with
  • 3
  • 3
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Métodos de diagnóstico em modelos logísticos trinomiais / Methods of dignóstics in trinomials Logistic models

Silva, Jose Alberto Pereira da 10 October 2003 (has links)
Os modelos logísticos trinomiais podem ser interpretados como uma extensão natural do modelo logístico binomial para situações em que a resposta admite apenas três resultados. Introduzimos inicialmente os modelos logísticos trinomiais e discutiremos em seguida alguns aspectos inferenciais, tais como estimação e testes. Medidas de qualidade do ajuste são também apresentadas. Contudo, o principal foco deste trabalho é a apresentação de métodos de diagnóstico. Mostramos que as técnicas usuais de diagnóstico desenvolvidas para o modelo logístico binomial podem ser adaptadas para o caso trinomial. O desenvolvimento de métodos diretos para o modelo logístico trinomial é mais complexo do ponto de vista computacional, embora seja sempre possível. Discutimos alguns desses métodos, dentre os quais, o desenvolvimento de resíduos, de métodos para detectar pontos de alavanca, métodos de deleção de pontos e influência local. Comparamos os métodos adaptados com alguns métodos diretos através de exemplos. / The trinomial logistic models can be interpreted as a natural extension of the traditional binomial logistic model to situations in which the response allows only three possible results. We firts introduce the trinomial logistic modles and then some inferential aspects, such as estimation and hypothesis testing are discussed. Good-of-fit measures are also given. However, the aim of this work is the presentation of some diagnostic procedures for trinomial logistic models. We show that methods developed for binomial logistic models can adapted straightforward for trinomial models. The developement of direct methods, even though possible, in general requires more complex calculation. Some of these direct methods, suchs as evaluation of residuals, measures of high leverage points, deletion methods and local influence are apresented. Coparisons between adapted and direct methods are made via examples with real data.
2

Métodos de diagnóstico em modelos logísticos trinomiais / Methods of dignóstics in trinomials Logistic models

Jose Alberto Pereira da Silva 10 October 2003 (has links)
Os modelos logísticos trinomiais podem ser interpretados como uma extensão natural do modelo logístico binomial para situações em que a resposta admite apenas três resultados. Introduzimos inicialmente os modelos logísticos trinomiais e discutiremos em seguida alguns aspectos inferenciais, tais como estimação e testes. Medidas de qualidade do ajuste são também apresentadas. Contudo, o principal foco deste trabalho é a apresentação de métodos de diagnóstico. Mostramos que as técnicas usuais de diagnóstico desenvolvidas para o modelo logístico binomial podem ser adaptadas para o caso trinomial. O desenvolvimento de métodos diretos para o modelo logístico trinomial é mais complexo do ponto de vista computacional, embora seja sempre possível. Discutimos alguns desses métodos, dentre os quais, o desenvolvimento de resíduos, de métodos para detectar pontos de alavanca, métodos de deleção de pontos e influência local. Comparamos os métodos adaptados com alguns métodos diretos através de exemplos. / The trinomial logistic models can be interpreted as a natural extension of the traditional binomial logistic model to situations in which the response allows only three possible results. We firts introduce the trinomial logistic modles and then some inferential aspects, such as estimation and hypothesis testing are discussed. Good-of-fit measures are also given. However, the aim of this work is the presentation of some diagnostic procedures for trinomial logistic models. We show that methods developed for binomial logistic models can adapted straightforward for trinomial models. The developement of direct methods, even though possible, in general requires more complex calculation. Some of these direct methods, suchs as evaluation of residuals, measures of high leverage points, deletion methods and local influence are apresented. Coparisons between adapted and direct methods are made via examples with real data.
3

Analýza chybějících hodnot: porovnání vhodnosti tradičních metod napříč mechanismy / Analysis of Missing Data: Comparing Performance of Traditional Methods across Mechanisms

Petrúšek, Ivan January 2014 (has links)
The objective of this thesis is to evaluate different methods of dealing with missing values in data analysis. The thesis is divided into three major chapters. The first chapter summarizes the theoretical literature on missing data and focuses on missing data mechanisms in particular. The second chapter introduces traditional methods for addressing missing data in sociological research. The third chapter assesses the performance of these methods by analyzing simulated data sets for two variables (income, IQ). For practical analysis (chapter 3), we simulated missing data according to three different mechanisms (MCAR, MAR, NMAR) and varied the proportion of missing values under these mechanisms (10%, 20%, 30%). Then, we applied each of the following four methods of addressing missing values: complete-case analysis, arithmetic mean imputation, regression imputation, and stochastic regression imputation. In order to evaluate the performance of each of these methods we performed correlation and regression analyses for each experimental condition. The results of these simulations are largely in agreement with existing theoretical literature on the subject of missing data. In the case of NMAR, all solution methods provided biased parameter estimates. In the case of MCAR, only complete-case analysis and...

Page generated in 0.0892 seconds