NLINFIT error of model being overparameterized

I need to fit some data using nlinfit but code is showing following error. May I please some some suggestion what can be specifically tried.Files attached.

Warning: The model is overparameterized, and model parameters are not identifiable. You will not be able to

compute confidence or prediction intervals, and you should use caution in making predictions.

> In nlinfit (line 377)

In Water_HIfit (line 16)

Answer 1

[John D'Errico]

When you learn your model is overparameterized, there is no magical solution. You can do one of two things:

1. Use a simpler model. Your model is too complex to be estimated from this data.
2. Get better data. Effectively, that means MORE data, but better data is always helpful in such a case.

This is not something you can fix by using some other method, or by flipping some switch.

[GAGANDEEP KAUR]

Actually, I am in doubt about its processing as I need to obtain the fitted values of taua1 and it is giving these values in in return along with this error message.

[John D'Errico]

You should be in doubt... But the fault lies in what you gave the software, not in the software.

The softwre told you that the data you provide is insufficient to estimate the model you want to fit. As such, anything it returns is suspect. To solve the problem, you need to do one of two things, as I said. Get better/more data, or a simpler model. I did look at the file you included. I saw a model that is wildly complex, so I would not try to suggest if it is even mathematically correct.

If I had to guess, it looks as if the code in there seems to be over-reaching your skills in MATLAB. It is as if you have jumped into the deep end of the pool without a clue how to swim. In your code, I see dozens of numbered variables, created for no good reason. Instead you need to learn to use MATLAB as it is designed to be used, as a vector and array tool. At the same time you need to learn concepts of nonlinear estimation and modeling, before you tackle the hard probems, learn to solve smaller ones.

I'm sorry, but you need to take a step or two backwards.

[GAGANDEEP KAUR]

Thank you for your response