Non-Parametric Analysis
To build a non-parametric model using a given data set:
| 1 |
Load the data and filter out the outliers either manually or automatically. |
- For details on eliminating outliers through data filtering, refer to Filtering Data.
| 2 |
Transform the data if a given parameter naturally ranges over more than one decade of values (e.g., 10-100). |
- For details on transforming data, refer to Transforming Data.
| 3 |
Choose Analysis > Non-Parametric Analysis to display the Non-Parametric Analysis dialog box. |
| 4 |
Enter the desired number of boundary points and the percentage of data to be enclosed. |
- For background details on modifying these options, refer to Using the Non-Parametric Analysis Dialog Box.
| 6 |
Click Check and Adjust Inputs to check and adjust the inputs for boundary points, percent enclosed, and diversity oversampling. |
| 7 |
Click OK to automatically analyze and display the results in a separate window. |
- Click Apply instead of OK to repeat the process and perform an analysis using different settings.
|
