Computational Example Part II Model Validation

Continuing with the orifice flow example from Sect. 3.3.4, model validation will now be performed by examining the residuals and the model quality and taking appropriate corrective action. 

The normal probability plot of the residuals is shown in Fig. 3.3. There seem to be some mild deviations from normality in the central region. Overall, given the small sample, there is not much that can be concluded with this particular sample. 

In order to determine model adequacy, aU the previous computed data will be collected and then examined. The values are  

There seems to be a single outlier at the point h = 0.10 m. Influential analysis, using Cook s distance is given below. Cook s distance for each of the points is given in Table 3.5. It can be seen that none of the points is individually influential as all the values are less than one. 

It can be noted that repeating the analysis by removing the identified outlier does not improve the fit or parameter estimates. Since the value of is high, the F-test is passed, and the confidence interval is small, it can be concluded that the model adequately describes the data and that the value of the resistance coefficient is approximately 28.4 kg min  

---