User Tools

Site Tools


en:data_preparation_r

Preparation of data for analysis

R functions

Simple EDA (Exploratory Data Analysis):

  • boxplot - draws the boxplot of data; with argument outline = TRUE (default setting) draws also the outliers;
  • hist - histogram with distribution of values in the variable;
  • qqnorm, qqline - draws QQ plot (comparing the quantiles of values in variable x to quantiles of normal distribution) and adds the line for comparison.
  • pairs - draws the pairwise relationships between variables in the matrix/data.frame; the original pairs function can be extended for panel functions from Numerical Ecology with R, which gives quite interesting tool for data overview (see details here).

Data transformation

  • sqrt
  • log, log10, log1p

Data standardization

  • scale
  • decostand (library vegan)
en/data_preparation_r.txt · Last modified: 2019/02/22 18:00 by David Zelený