How correlated are the variables in the data?

Create a correlation matrix of the selected variables. Correlations and p.values are provided for each variable pair. To show only those correlations above a certain (absolute) level, use the correlation cutoff box.

Note: Correlations can be calculated for variables of type numeric, integer, and date. Variables of other types with no more than two unique levels are transformed into 0-1 dummies and can also be selected. For these variable types, the first level is converted to a 1 and the second to a 0. A visual representation of the correlation matrix is provided in the Plot tab. Note that scatter plots in the graph at most 1,000 data points by default. To generate scatter plots that use all observations use plot(result, n = -1) in Report > Rmd.

Stars shown in the Plot tab are interpreted as:

• p.value between 0 and 0.001: ***
• p.value between 0.001 and 0.01: **
• p.value between 0.01 and 0.05: *
• p.value between 0.05 and 0.1: . The font-size used in the plot is proportional to the size and significance of the correlation between two variables.

### Method

Select the method to use to calculate correlations. The most common method is Pearson. See Wikipedia for details.

### Correlation cutoff

To show only correlations above a certain value choose a non-zero value in the numeric input between 0 and 1 (e.g., 0.15).

### Covariance matrix

Although we generally use the correlation matrix, you can also show the covariance matrix by checking the Show covariance matrix box.

### Report > Rmd

Add code to Report > Rmd to (re)create the analysis by clicking the icon on the bottom left of your screen or by pressing ALT-enter on your keyboard.

By default the correlation plot samples 1,000 data points. To include all data points use plot(result, n = -1) To add, for example, a title to the plot use title(main = "Correlation plot\n\n"). See the R graphics documentation for additional information.

### R-functions

For an overview of related R-functions used by Radiant to evaluate correlations see Basics > Tables

© Vincent Nijs (2018) 