Test Groupings

The Define Test Groupings dialog collects records of the current data object into groups; these groups are referred to as tests.

The dialog is accessed from the Data Editor in either of these ways:

• Using the menu Tools > Change Test Groupings

• Using the toolbar button

When you enter the dialog, a plot is displayed as the variable logno is automatically selected for grouping tests.

Select another variable to use in defining groups within the data.

1. Select n in the Variables list.

2. Click the    button to add the variable (or double-click n).

   The variable n appears in the list view on the left as seen in the following example. You can now use this variable to define groups in the data. The maximum and minimum values of n are displayed. The Tolerance is used to define groups: on reading through the data, when the value of n changes by more than the tolerance, a new group is defined. You change the Tolerance by typing directly in the edit box.

   You can define additional groups by selecting another variable and choosing a tolerance. Data records are then grouped by n or by this additional variable changing outside their tolerances.

3. Clear the box Group by for logno. Notice that variables can be plotted without being used to define groups.

4. Add load to the list by selecting it on the right and clicking  .

5. Change the load tolerance to 0.01 and watch the test grouping change in the plot.

6. Clear the Group By check box for load. Now this variable is plotted without being used to define groups.

   The plot shows the scaled values of all variables in the list view (the color of the tolerance text corresponds to the color of data points in the plot). Vertical pink bars show the tests (groups). You can zoom the plot by Shift-click-dragging or middle-click-dragging the mouse; zoom out again by double-clicking.

7. Select load in the list view (it becomes highlighted in blue) and remove it from the list by clicking the  button.

8. Double-click to add spark to the list, and clear the Group By check box. Select logno as the only grouping variable.

   It can be helpful to plot the local model variable (in this case spark) to check you have the correct test groupings, as shown below. The plot shows the sweeps of spark values in each test while speed (n) is kept constant. Speed is only changed between tests, so it is a global variable. Try zooming in on the plot to inspect the test groups; double-click to reset.

   

   Reorder records allows records in the data set to be reordered before grouping. Otherwise, the groups are defined using the order of records in the original data object.

   Show original displays the original test groupings if any were defined.

   One test/record defines one test per record, regardless of any other grouping. This is required if the data is to be used in creating one-stage models.

   Test number variable contains a pop-up menu showing all the variables in the current data set. Any of these could be selected to number the tests.

9. Make sure logno is selected for the Test number variable.

   This changes how the tests are displayed in the rest of the Model Browser. Test number can be a useful variable for identifying individual tests in Model Browser and Data Editor views (instead of 1,2,3...) if the data was taken in numbered tests and you want access to that information during modeling.

   If you chose none from the Test number variable list, the tests would be numbered 1,2,3 and so on in the order in which the records appear in the data file. With logno chosen, you will see tests in the Data Editor listed as 586, 587 etc.

   Every record in a test must share the same test number to identify it, so when you are using a variable to number tests, the value of that variable is taken in the first record in each test.

   Test numbers must be unique, so if any values in the chosen variable are the same, they are assigned new test numbers for the purposes of modeling (this does not change the underlying data, which retains the correct test number or other variable).

10. Click OK to accept the test groupings defined and dismiss the dialog.

    You return to the Data Editor window. At the top is a summary of this data set now that your new variable has been added and a new filter applied (example shown below).

    

11. The number of records shows the number of values left (after filtration) of each variable in this data set, followed by the original number of records. The color coded bars also display the number of records removed as a proportion of the total number. The values are collected into a number of tests; this number is also displayed. The variables show the original number of variables plus user-defined variables.

Storage

Matching Data to Designs

© 1984-2012- The MathWorks, Inc.    -   Site Help   -   Patents   -   Trademarks   -   Privacy Policy   -   Preventing Piracy   -   RSS