How to determine the initial values for the regression correlation parameters in Aspen?
Thread Content
When I use experimental data (binary or multivariate) for data regression and choose a property-based method, what should I do if Aspen does not provide initial values for the model parameters? Can it be estimated using the material structure? Since I’m a beginner, I don’t quite understand. Please, dear friends, help me out. Thank you!
It’s fine to set the initial value to zero; you’re probably referring to the initial value used in regression. It shouldn’t affect the regression results, only the number of iteration steps required for the regression