How much a model with:
- Model p-value < 0.05
- Lack of fit p-value > 0.1
- R2=0.99, Adj. R2= 0.99, Pred. R2=0.99
- normal probability plot attached (residuals do not follow a straight line).
is reliable?
Impossible to anser. You will have to specify what model you mean (structual and random part). And further, you should have some criteion of "reliability". How would you measure it? What are acceptable values for the "reliability"?
The values you gave are quite meaningless. The p-value refers to some null hypothesis of a specified model. If the model is miss-specified, neither the null nor the p-value provide any reasonable insight. The same applies to the lack of fit. The R² values were not useful even when the model is correctly specified, since they give only the proportion of explained variance, what is not related to reliability.
Showing "R²" and "Normal probability plot" implies that the random component of your model is based on the normal distribution. The plots indicate that the underlying assumptions are far from being reasonable in your case.
Hello Hamed,
I fully agree with Jochen, however I have some questions from you.
Did you use design of expert program for your purpose?
If so, this is happening because of your level of combination effects that you introduced to the model. How many combination of your parameters were included in your design (all?)? try to decrease the level and see the results. Meanwhile, you can use Pareto Chart for understanding the level of combination of your parameters and then building model based on the observed influential combinations. Ignoring this plot is also another source of obtaining the results just like yours!
Please, let me know if the above recommendations were helpful for your case.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis