SAS/INSIGHT software is a highly interactive tool for data analysis. You can explore data through a variety of interactive graphs that include bar charts, scatter plots, box plots, and three-dimensional rotating plots. You can examine distributions and perform parametric and nonparametric regression, analyze general linear models and generalized linear models, examine correlation matrixes, and perform principal component analyses. Any changes you make to your data show immediately in all graphs and analyses. You can also configure SAS/INSIGHT software to produce graphs and analyses tailored to the way you work.

SAS/INSIGHT software is an integral part of the SAS System. You can use it to examine output from a SAS procedure, and you can use any SAS procedure to analyze results from SAS/INSIGHT software.

SAS/INSIGHT software includes features for both displaying and analyzing data interactively. A data window displays a SAS data set as a table in which the columns of the table display variables and the rows display observations. Data windows provide data management features for editing, transforming, subsetting, and sorting data. A graph window displays different types of graphs: bar charts, scatter plots, box plots, and rotating plots. Graph windows provide interactive exploratory techniques such as data brushing and highlighting. Analysis windows display statistical analyses in the form of graphs and tables. Analysis windows include the following features:

• univariate statistics

• robust estimates

• density estimates

• cumulative distribution functions

• theoretical quantile-quantile plots

• multiple regression analysis with numerous diagnostic capabilities

• general linear models

• generalized linear models

• smoothing spline estimates

• kernel density estimates

• correlations

• principal components

SAS/INSIGHT software might be of interest to users of SAS High-Performance Forecasting software for interactive graphical viewing of data, editing data, exploratory data analysis, and checking distributional assumptions.