Hi, I am using simplex centroid design for analyzing the interactions between three variables, some of the responses are fitting with Linear and some fit with Quadratic models, so I was wondering if it is possible to I explain in the paper that responses A, B, and C were fitted with Linear model and responses D, E and F were fitted with Quadratic? or I should use only one model in the manuscript? I only used quadratic but reviewer asked me to check linear to fit some of them (which he is correct and they are linear) but now i am wondering can i report them as linear or just remove them from the manuscript?
It is up to you but it is recommend to use SAS program it is really helpful in your case
Thanks for the answer, I use Minitab and Design expert for both analyzing and drawing the contour plots.
It is possible if the data are well collected. Check if your p values are significant for all the models developed and use 20% of your data for model validation..
Hi dear Diako
You should propose a model, but I think the reviewer realized from your data that you made the mistake to designing the data or proposed model is over fitted.
Dear @Diako Javanmardi, model is viewed as an input-output transformation for these tests. The validation test consists of comparing outputs from the system under consideration to model outputs for the same set of input conditions (I mean here 20% of data that was not infused in model development). Data recorded while observing the system must be available in order to perform this test. The model output that is of primary interest should be used as the measure of performance. For example, if system under consideration is a fast food drive through where input to model is customer arrival time and the output measure of performance is average customer time in line, then the actual arrival time and time spent in line for customers at the drive through would be recorded. The model would be run with the actual arrival times and the model average time in line would be compared with the actual average time spent in line using one or more tests.
I believe there is a reason that the reviewer asked you to check linear model. There are 2 options:
1- Use linear model for the data and add it to the manuscript and mention it in the methodology.
2- In the methodology mention that you will only check it with quadratic model; then, in the conclusion suggest checking the data with linear model for future research and studies.
Hope it helps
You are using a simple centroid design, so that means that you have data for a surface response; which includes a linear, quadratic and interaction for three factors or more if you have it. Analysis can be done by using SAS. Check surface response design which some authors call star design, which is not an appropriate form.
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling