Moving from nlinfit to lsqnonlin?

I want to estimate the parameters of a nonlinear model. Using nlinfit so far works, and makes sense to me. I use it in the following way:

b=nlinfit(x_data,y_data,@my_model,initial_guess);

However, I want to add constraints to the parameters I'm estimating, which is why I'm trying to switch to lsqnonlin. I'm having a hard time translating my problem to this function though. After looking at the documentation, I can't see how I can pass the x and y data using this function. I literally just need what nlinfit does except with constraints to the parameters. Do I need to somehow make the "my_model" function accept the x and y data instead? Any help would be appreciated.

[John D'Errico]

lsqnonlin does not allow general constraints. Only bound constraints.

Answer 1

[Sean de Wolski]

lsqcurvefit is essentially the same thing as lsqnonlin but accepts data and functions in the same manner you're used to.

[John Alperto]

OK, let me give this a shot.