Hi every one!
Do you know that what is this error in General Linear Model in Minitab 20?
''The following terms cannot be estimated and were removed: A*B, A*C, B*C''
Why sometimes Minitab can't calculate some interactions of parameters?
And how can I fix this problem?
because I need to calculate the P-value of interaction of this parameter!
The study here is conducted by Taguchi L9 orthogonal array, which is selected by 3 levels of design with 3 factors.
A B C
150 2 300
150 4 600
150 6 900
200 2 600
200 4 900
200 6 300
250 2 900
250 4 300
250 6 600
studying the machinability of a material ..
Hello Amr,
Without additional information about the nature of your variables (e.g., are IVs categorical or continuous?), data set, and intended research question(s), all I can do is offer some general suggestions as to why effects which you believe could (and should) be estimated are not in a linear model:
1. You have missing cell(s) or observations for one or more specific combinations of terms;
2. You have inadequate observations per cell to estimate the desired terms (when IVs are categorical);
3. Terms are collinear (with one or more other terms in model);
4. Data cases may be mislabeled on unidentified as to their appropriate value/level on one or more variables;
Good luck with your work.
thank you for your response to my question.
The study here is conducted by Taguchi L9 orthogonal array, which is selected by 3 levels of design with 3 factors.
A B C
150 2 300
150 4 600
150 6 900
200 2 600
200 4 900
200 6 300
250 2 900
250 4 300
250 6 600
studying the machinability of a material ..
David Morse
Amr, please note that the L9 method (at least in its original format) is suitable exclusively to study the main effects but not their interactions. In order to overcome this shortcoming, you have to specify the interaction terms a priori and assign them to the column in the orthogonal array.
- Badges
- Science topic