simultaneous fitting with one shared parameter

[photoon]

I can try to do the similar algorithm of your vi, but I cannot see the block diagram of your program. Could you teach me how to do that?

[DSPGuy]

Christian's example is excellent.  To help you get started I am attaching a simple example that may help.  The model is the sum of two exponential functions.  Each exponential has the form a*exp(b*x)+c, and the b parameter is shared across two data sets.  a and c are different for each of the two exponentials.

-Jim

[photoon]

I briefly looked at the example and it seems that this will be a great way to follow. I will work on and let you people know when I make more idea. 

Thanks

-Doogie

[photoon]

DSPGuy,

I can understand your example except one thing. It seems to me that 2 other parameters are fixed, not really fitting parameters. Is it true? Or if they are fitting parameters, how can I get those returned values after fitting?

[altenbach · Accepted solution]

No, there are currently no other fitting paramters in Jim's example, partially because your original description is a bit confusing.

For example, you said "But, other three parameters are not supposed to be matched.", which could be understood that you don't want to fit for them at all. However, you meant that they should be different for each set (i.e. not shared), but still to be fit.

To rewrite for two unique paramters for each set and on shared parameter, you would rewrite it as follows, now with 5 parameters.

You should be easily able to adapt it to your problem.

 

[photoon]

Christian and Jim,

Now I completely understand your example. 

I really appreciate your solution.

Best,

-Doogie