which statistical model is suitable to compare two curves for validation purposes, when comparing two force-displacement curves for instance?
It is suggested to perform Correlation Analysis Between numerical and experimental result on the one hand, on the other hand you can compare the maximum values of the curves.
I also have this question, how to make sure the agreement between experimental and numerical methods is good or not? Some researchers just compare the maximum values
You may perform statistical significant test for that. For example- student t-test.
Cheers!
Hi Ali.
this answer, necessarily has 3 elements to consider, first and in order, what is the research question?, second which research design did you use? Are paired measurements? and third, what do you expect to say?
in the best of cases, the majority of the colleagues who state that it is a possibility, the regression, that in the light of what it explains, makes more sense.
But if what you want to explain goes a little further, such as: behavior, variability, asymmetry, inflection points, trends, this could be implicit in the regression, as a general result, but not necessarily seen.
I'm going to speculate what you want to find, "is it possible that the experimental values estimate the behavior of the measured values?", regression is ok.
If the question is different, then consider adding another kind of analysis.
Dear Ali,
You can use Mean Absolute Percentage Error (MAPE) method to comparison between experimental and predicted data.
Dear Ali,
You can use Mean Absolute Percentage Error (MAPE) method to comparison between experimental and predicted data.
You can use the design of experiments by YATES or TAGUCHI method for example
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling