Visualization of Skewed Data: A Tool in R

After discussing the main characteristics of the histogram and of a number of variations in the boxplot, this work presents a visualization tool specifically tailored to deal with skewed data. The idea is to use various types of boxplots (the classical one, which is tuned for skewness of the data, the shifting boxplot, and the box-percentile plot), the violin plot, and the histogram with a nonparametric estimate of the density overlay. The plots are presented in such a way that they facilitate the extraction of additional information from each one. We show that a good deal of information can be extracted from the inspection of the output using example data from synthetic aperture radar images. We provide an implementation in R based on functions already available.

• Autores: Ospina, Raydonal; Larangeiras, Antonio Marcos; Frery, Alejandro C.
• Tipo de recurso: Article of journal
• Fecha de publicación: 2014
• Institución: Universidad Nacional de Colombia
• Repositorio: Universidad Nacional de Colombia
• Idioma: spa
• OAI Identifier: oai:repositorio.unal.edu.co:unal/66561
• Acceso en línea: https://repositorio.unal.edu.co/handle/unal/66561; http://bdigital.unal.edu.co/67589/
• Palabra clave: 51 Matemáticas / Mathematics; 31 Colecciones de estadística general / Statistics; Exploratory Data Analysis; Skewed Data; Boxplot; Violin Plot; Visualization; Análisis exploratorio de datos; Boxplot; Datos sesgados gráficos de violín; Visualización.
• Rights: openAccess
• License: Atribución-NoComercial 4.0 Internacional