lsqnonlin question

Consider the following code:

X(1:10304)=xlsread('qs_comp_2.xls',2,'G3822:G14125');
Y(1:10304)=xlsread('qs_comp_2.xls',2,'H3822:H14125');

X(10305:12631)=xlsread('comp.xls',3,'T4259:T6585');
Y(10305:12631)=xlsread('comp.xls',3,'U4259:U6585');

X0=[1012 1400 0.17 -0.0001];
lb = [1012;900;0.17;-0.0001];
ub=[1300;1600;0.3119;10];
StartAt =  [1012;900;0.17;-0.0001];
 options = optimset('MaxFunEvals',10000);
x=lsqnonlin(@(X0)fit_simp(X0,X,Y),StartAt,lb,ub,options);

[total_readings,epsilon_dot_QS,epsilon_dot_MR] = GetMRDetails;

for i=1:total_readings

         Y_new(i)=(X0(1)+X0(2)*(X(i)^X0(3))+X0(4)*log(epsilon_dot_QS/epsilon_dot_MR));

end

Please can anyone help me here:

1) Is the final optimised vector X0 or is it x? That is: which id the final answer X0 or X 2)Is there anything I'm missing in this code?

Answer 1

[Sean de Wolski]

It is x, the output from lsqnonlin. It doesn't look like the above code ever uses the actual optimized value.

[Shalini]

Please can you correct it-I shall be grateful.I'm new to matlab

Answer 2

[Shalini]

Why you think so?

My fit_simp function (see properly above) is as below

function diff = fit_simp(x,X,Y)

% This function is called by lsqnonlin.
% x is a vector which contains the coefficients of the
% equation.  X and Y are the option data sets that were
% passed to lsqnonlin.

A=x(1);
B=x(2);
n=x(3);
C=x(4);

[total_readings,epsilon_dot_QS,epsilon_dot_MR] = GetMRDetails;

     for i=1:total_readings

          diff(i)=(A+B*(X(i)^n)+C*log(epsilon_dot_QS/epsilon_dot_MR));

          diff(i)=diff(i)-Y(i);

      end

[Sean de Wolski]

That looks fine, it's that your call to LSQNONLIN generates a vector called 'x'. This vector is not used in the remainder of the code you showed us.

[Shalini]

ok, i will correct that- 2 questions:

1)IS optimisation code correct?

2) I want to outtput the adjusted R square, how to do that?

Shalini