Packages

In R, there are several packages that provide the correlation functionality to the programmer. We will be using the following packages in this chapter:

• corrgram: This is the tool to graphically display correlations

• Hmisc: This contains a variety of miscellaneous R functions

• polycor: This contains functions to compute polychoric correlations

• ggm: This contains functions for analyzing and fitting graphical Markov models

Correlation

Basic correlation is performed in R using the cor function. The cor function is defined as follows:

cor(x, y = NULL, 
  use = "everything", 
  method = c("pearson", "kendall", "spearman"))

The various parameters of this function are described in the following table:

Parameter	Description
x	This is the dataset.
y	This is the dataset that is compatible with x.
use	This is the optional method for computing the covariance of missing values assigned. The choices are: everything all.obs complete.obs na.or.complete pairwise.complete.obs
method...