IRI Home

CPT Help Home -> Specific Menu Items -> Tools -> Regression

Tools Regression

If Multiple Linear Regression (MLR) is used, the Tools ~ Regression menu item opens a child window showing the regression equation(s) for estimating the value of the Y variable using the X variable(s). If there is only one X variable, a graph will also be shown comparing the predictor with the predictand. This graph is different to that shown in the Scatter Plot: the scatter plot compares the cross-validated predictions with the observations, whereas the Regression Plot compares the predictors with the observations.

The equations can be saved by selecting the MLR Results tab from File ~ Output Results. The y-axis limits on the regression plot can be set by right-clicking on the image and then selecting the Customise ~ Y-Axis Limits option. The maximum and minimum values on the y-axis are indicated and can be reset. Resetting these axis limits will in turn automatically reset the tick-mark interval and number of tick marks. Greater flexibility for setting the tick marks is planned for future releases. Further options for customizing the graphs, including adjusting the size of the fonts, and the size of the images, are available from Options ~ Graphics.

The graph can be saved individually as a BMP, EMF, GIF, JPEG, PCX, or PNG file by right-clicking anywhere in the child window, and then selecting Export. A default name is given to the graphic file, but this name can be changed using the browse button. The quality of the JPEG file can be adjusted using the slider or by the quality indicator, which ranges between 0.01 and 1.00. The highest quality is obtained using 1.00. The size of the JPEG file is affected by the quality chosen, with larger files being generated the higher the selected quality. The image quality for all the graphics formats can be improved by increasing the size of the image (see Options ~ Graphics ~ Graph Scaling). The graphic title can be set using the Customise option upon right-clicking in the child window.


 
Last modified: