This procedure performs regression with a single predictor, in linear, second-order or third-order polynomial terms.
1. Choose ProcessMA > Statistics > Regression > Regression (Single Predictor)
2. In Response (Y), select the column containing the response data.
3. In Predictor (X), select the column containing the predictor data.
4. In Model type, select linear, quadratic or cubic model and
5. Click OK.
Optional
6. Check Plot fitted line, if you want to display a histogram of the data.
7. Check Show confidence bands, if you want to display the confidence bands in the fitted line plot.
8. Check Show prediction bands, if you want to display the prediction bands in the fitted line plot.
9. In Confidence level, enter the desired confidence level.
10. Check Plot residuals, if you want to display the residual plots for analysis.
Note To select a column of data into a textbox, double-click on any of the column names shown in the list on the left of the dialog box while in the textbox.
Response (Y): Numeric.
Predictor (X): Numeric.
Confidence level: Numeric; Between 0 to 100.
You want to model the relationship between the size of the valve and the speed of the machine using a quadratic model. You collected data on various settings of the valve size and the resultant speed of the machine.
1. Open worksheet ProcessMA > Tools > Data Files > Regression.xls.
2. Choose ProcessMA > Statistics > Regression > Regression (Single Predictor).
3. In Response (Y), select A – Speed.
4. In Predictor (X), select B – Valve.
5. In Model type, choose Quadratic.
6. Check Plot fitted line.
7. Check Show confidence bands.
8. Check Show prediction bands.
9. In Confidence level, enter 95.
10. Click OK.
1.gif)
Interpretation
The P-value for the model suggests a good fit to the data. The R2 value indicates that about 94% of the variation in speed is explained by the variation in the valve size. The 95%CL lines are the confidence limits for the machine speed and the 95% PL lines are the prediction limits for new observations.