norm(z-y');

function err = fitlognormal(lambda,t,y)
%	Fitting function for lognormal, lambda(1)=position, lambda(2)=width
%	between the data and the values computed by the current
%	function of lambda.  Fitlognormal assumes a lognormal function 
%  T. C. O'Haver, May 2006
global c
A = zeros(length(t),round(length(lambda)/2));
for j = 1:length(lambda)/2,
    A(:,j) = lognormal(t,lambda(2*j-1),lambda(2*j))';
end
c = A\y';
z = A*c;
err = norm(z-y');