How to Perform a Cubic Regression - dummies

How to Perform a Cubic Regression

By Jeff McCalla, Steve Ouellette

If a cubic polynomial is a good fit for the data on a scatter plot, you need to perform a cubic regression. You can do this in the TI-Nspire Lists & Spreadsheets application. To perform a regression, follow these steps:

  1. Press


    to move back to the Lists & Spreadsheet page containing the data needed.

  2. Press [MENU]→Statistics→Stat Calculations→Cubic Regression.

    A dialog box opens. As with any dialog box, you can press [TAB] to move from one field to the next or [SHIFT][TAB] to move backward through the fields. Press [ENTER] at any time for the configuration to take effect.

    Notice in the first screen that pressing [CLICK] and selecting year enables you to specify the location of the X List. You could also specify the column by its letter reference. In the second screen, pressing [D][CTRL][ ( ] enables you to indicate where you want the results to be stored. Column D because it is the first available empty column.


    The third and fourth screens (with columns D and E widened) show that the results of the regression are pasted directly into the spreadsheet. Notice that the coefficient of determination, R2, is very close to 1. This might suggest that the choice of doing a cubic regression was a good one.

A cubic regression can be done with a minimum of four points. Under this scenario, TI-Nspire finds a third-degree polynomial fit by solving a system of four equations determined by using the x and y-values of the four points. With five or more points, TI-Nspire performs a cubic regression.