The treatment of variables in the Distribution platform is different, depending on the modeling type of the variable, which can be categorical (nominal or ordinal) or continuous.
For categorical variables, the initial graph that appears is a histogram. The histogram shows a bar for each level of the ordinal or nominal variable. You can also add a divided (mosaic) bar chart.
The Frequencies report show counts and proportions. You can add confidence intervals and test the probabilities from the options in the red triangle menu.
For numeric continuous variables, the initial graphs show a histogram and an outlier box plot. The histogram shows a bar for grouped values of the continuous variable. The following options are also available:
• normal quantile plot
• quantile box plot
• stem and leaf plot
• CDF plot
The reports show selected quantiles and summary statistics. Additional report options are available in the red triangle menu for the following:
• saving ranks, probability scores, normal quantile values, and so on, as new columns in the data table (Save options)
• testing the mean and standard deviation of the column against a constant you specify (Test Mean and Test Std Dev options)
• fitting various distributions and nonparametric smoothing curves (Continuous Fit and Discrete Fit options)
• performing a process capability analysis for a quality control application
• confidence intervals, prediction intervals, and tolerance intervals