ill-conditioned of The Jacobian

Hi,

First of all, i'm sorry for my english if some of my sentences wont be clear, I'm a french student.

I'm working on a exercice which ask me to use the command "nlinfit". I did all the exercice, but when i run it, i have this message which appears : "Warning: The Jacobian at the solution is ill-conditioned, and some model parameters may not

be estimated well (they are not identifiable). Use caution in making predictions. ".

I already saw that this question has already been asked, but i cant solve it with my code. I dont see where the problem is. Can someone help me ?

I dont know if i can show my code, but here it is :

I would be very grateful if someone can help me to find where the problem is.

load('data.mat'); 
n=nlinfit(x,y,@parabole,[-10.^(-5);-1;10.^4]); 
n1=n(1); 
n2=n(1);
n3=n(3);

p=n1.*x.^2+n2.*x+n3; 
e=[((-n2-sqrt(n2^2-4*n1*n3))/2*n1);((-n2+sqrt(n2^2-4*n1*n3))/2*n1)];
x1=e(1); 
x2=e(2); 

hold on 
if x1>0 %Si x1 positif
    plot(x1,0,'o'); 
end

if x2>0 %Si x2 positif
    plot(x2,0,'o'); 
end

plot(x,y,'m');
plot(x,p,'g'); 
hold off 

function[p]=parabole(a,x) 
a1=a(1);
a2=a(2);
a3=a(3);
p=a1+a2.*x+a3.*x.^2;
end 

Thank you very much

[Jan]

What is the contents of the file data.mat?

By the way, -10.^(-5) is an expensove power operation, while -1e-5 is a cheap constant.