One could suppose that for some variables in social sciences and psychology could be an optimal level as well as in ecology and biology, so we could find a nonlinear correlation between variables.
Looks like exists a bunch of such methods and even linear correlation could someway catch nonlinear dependencies https://goldinlocks.github.io/Comparing-Linear-and-Nonlinear-Correlations/
Probably better nonlinear correlation coefficients are Distance correlation, Cosine similarity and Maximal information Coefficient.
Does exist some software for computing these coefficients and their p-levels by easy-way, without some programming?
Minitab v.19 do have the option to do nonlinear multiple regression. I would believe other off the shelf statistics packages can do something similar. I do not known with which version it came.
To some extent Excel can fit some nonlinar functions to a xy-dataset. There are options for exponential, logarithmical, polynomes up to the 6th degree, potential, and linear, of course. However only one pair at a time can be done. I would also believe there may exist statistics packages for Excel that can do multiple nonlinar regression. I have used a linear mulitiple regression package for Excel some 10-15 years ago.
You can try mine ("FittingKVdm"): simple to use and many examples included. Looking forward to hearing your comments/suggestions!
www.lerenisplezant.be/fitting.htm
origin can be used for various plot and matmatical analysis .various option are there
- Badges
- Science topic
- Similar topics
- Biological Science
- Physiology
- Physiological Processes
- Stress