Using SAS and R for visualizations
R is a software that can be integrated into reporting SAS data. With R, which is open source, it is possible to set up connections between SAS and R data. But the main difference between making plots and other visualizations in SAS versus doing it in R has to do with data handling. As we have seen with SAS, when using PROCs that create plots, such as PROC UNIVARIATE
, SAS typically reads or calculates the relevant values from the entire dataset and plots them. In a scatter plot, this is necessary – but it is not necessary for all plots. Although some SAS PROCs have the ability to take in a summary dataset and visualize it, many SAS PROCs require processing the whole underlying dataset.
Let's think of a box plot for a moment. For a box plot, outliers aside, we technically only need to know five different points in order to create the image of the plot: the minimum, 25th percentile, median, 75th percentile, and maximum. If we were creating...