Modeling dependency between two variables

Question

I want to model the dependence between two variables. Now there are a few ways to do this, for example writing one variable as a function of the other or using correlation. I have also looked into copulas but I was wondering if anyone knows any other way to model the dependence of two variables?

Help is much appreciated!

Answer 1

In the absence of a physical model, first have a scatter plot of your data. This could give you an idea of the type of model to work (linear, polynomial, exponential, logarithmic, ...). When you have selected the model, regression (linear or non linear) techniques will need to be used (least squares methods). If the physical model is already available, similar methods will allow you to fine tune the model parameters.

Answer 2

Look at the Iman and Conover IHQ method