Publication date: 04/12/2021

Explore Patterns Report

The initial Explore Patterns report contains a Control Panel, Univariate Summary report, and Univariate Patterns report. Many of the tables in these reports contain rows that can be selected. If a row is selected and you right-click the first column, a menu appears that enables you to select or colorize the related rows and columns in the data table.

Control Panel

The control panel contains options that enable you to select which reports appear in the report window. These are the same options that are in the Explore Patterns red triangle menu. See Explore Patterns Utility Options. There is also a section of the control panel to specify report settings. These are the same settings that are found in the launch window. See Launch the Explore Patterns Utility. If you change the settings in the control panel, the reports are automatically updated.

Univariate Summary

The Univariate Summary report combines results from several patterns reports across columns. By default, there are summary tables for Longest Runs and Longest Duplicate Sequences. Longest Runs and Longest Duplicate Sequences with rarity 2 are shown in the tables.

If the Fraction Length option is selected a Fraction Length summary table is also shown. This table shows the proportion of values in each column that have a fraction length greater than or equal to 15. Columns that do not contain any values that have fraction lengths greater than or equal to 15 are not shown in the Fraction Length summary table.

If at least one of the Y columns has a Spec Limit column property and the Distribution wrt Spec Limits option is selected, a Spec Limits Distrib summary table is also shown. This table shows the columns that have a statistically significant number of observations outside of the specification limits.

For more information about the Univariate Summary report contents, see Explore Patterns Utility Options.

Univariate Patterns

The Univariate Patterns report contains a Select Columns list and one or more column-specific patterns reports. The column-specific reports that are displayed are determined by the Select Columns list. To examine each column’s report, select the first column and then successively use the down arrow key to advance to the next column. Alternatively, you can click a column name in this list to display the patterns report for the selected column. If you select more than one column name by using either Ctrl or Shift, a patterns report for each of the selected columns is displayed. If you deselect one or more columns from the Select Columns list, the reports for the corresponding columns are removed.

Each individual patterns report displays a summary table that contains the number of rows, number of missing observations, and number of unique values for the given column. By default, there are also tables for Most Duplicated Values, Longest Runs, and Longest Duplicate Sequences. If the Formatted Widths, Fraction Lengths, or Leading and Trailing Digits options are selected from either the control panel or the Explore Patterns red triangle menu, a corresponding table is added to each individual patterns report.

If you right-click on a row in any of the tables in the Univariate Patterns report, a menu of right-click options appears.

Select Rows and Column

Selects the column and rows in the original data table that correspond to the selected row in a specific table in the Univariate Patterns report. For example, if you right-click the second row of the Leading Digits table for a specific column, the rows in that column that have a leading digit of 2 are selected in the original data table.

Colorize Cells

Colors the cells in the original data table that correspond to the selected row in a specific table in the Univariate Patterns report. The rows of the colored cells are also selected.

There are red triangle options for the column-specific patterns report. See Univariate Patterns Report Options.

Want more information? Have questions? Get answers in the JMP User Community (community.jmp.com).