Paquete sobre el Mathematica 8.0 para variantes del análisis de Regresión Categórica
Cargando...
Archivos
Fecha
Autores
Orama Gómez, Claudia
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Central “Marta Abreu” de Las Villas
Resumen
El análisis de datos categóricos es una rama importante de la estadística. Su campo de aplicación es amplio. Las ciencias sociales y las ciencias médicas son buenos ejemplos de ello.
Esta tesis trata sobre un método de análisis de datos categóricos: el análisis de regresión categórica. El trabajo comienza con un estudio exhaustivo del análisis de regresión múltiple y la verificación de los supuestos. Se describen numerosas pruebas gráficas y estadísticas para verificar la normalidad de los residuales y la homogeneidad de sus varianzas.
Se explican en detalle el modelo de regresión categórica y su algoritmo. Se creó y probó un paquete sobre el Mathematica 8.0 para realizar los cálculos fundamentales.
Se utilizaron datos simulados para mostrar y comentar los resultados más importantes. Además se presentó un problema con datos reales: el diagnóstico de la hipertensión arterial en la ciudad de Santa Clara.
The categorical data analysis is an important branch of Statistic. Their field of application is wide. Social Sciences and Medical Sciences are good examples. This thesis is about the method of analysis of categorical data: the categorical regression analysis. The work begins with an exhaustive study of the multiple regression analysis and the verification of the assumptions. Many graphical and statistical tests to verify the residual normality and the homogeneity of their variances are described. The model of the categorical regression analysis and its algorithm are carefully explained. A Mathematical package to do the fundamental calculus was created and proved. A simulated data was used in order to show and comment the most important results. Also a real data of a medical problem was presented: the diagnosed of the hypertension in the Santa Clara City.
The categorical data analysis is an important branch of Statistic. Their field of application is wide. Social Sciences and Medical Sciences are good examples. This thesis is about the method of analysis of categorical data: the categorical regression analysis. The work begins with an exhaustive study of the multiple regression analysis and the verification of the assumptions. Many graphical and statistical tests to verify the residual normality and the homogeneity of their variances are described. The model of the categorical regression analysis and its algorithm are carefully explained. A Mathematical package to do the fundamental calculus was created and proved. A simulated data was used in order to show and comment the most important results. Also a real data of a medical problem was presented: the diagnosed of the hypertension in the Santa Clara City.