nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

DOE results analysis

FR60

I was analyzing the results of a DOE with 5 factors moved up and down wrt set values. The response Y is a measured resistance.

I ran a fit model by using a full factorial and then a stepwise regression. The stopping rule was Min BIC and direction: forward. Once selected the suggested factors I made the model and ran it. The R2 was very good as also the profilers. Anyway if I look at residuals they are very strange. See below plot. The distance of each experiment value is always at the same distance wrt fitting line in absolute value. Why?

Thanks fr

Accepted answers

Mark_Bailey

I was guessing about your data and model, so I had to assume particular details that were not provided. It doesn't change anything: the result in the residual plot is due to limited DF for error and you have not done anything wrong.

All comments

Mark_Bailey

It is a matter of the degrees of freedom left for the error sum of squares. I suspect that if you check the Analysis of Variance outline in the Fit Least Squares platform, you will find that the DF for Error is 1. That constrains the residuals to be two levels. There is nothing wrong, you just don't have the independent information to really see the model errors. (Residuals from the model estimate the errors from the response.) If you remove higher order term, you will see the number of unique residuals levels double from 2 to 4.

FR60

Hi Mark

In the analysis of Variance the error DF is 2.

Felice

Mark_Bailey