Subscript indices must either be real positive integers or logicals. why does this come up?

Trial>> AusProfInt(2, 1, 1.5, 3, 100, 0.7)
Subscript indices must either be real positive integers or logicals.

Error in AusProfInt (line 8)
Ausdemand1= -(A*2^2)/2 + (A*((x^1/1-sigma)-1/pa-pc(x)^1/1-sigma)*(1 +
((x^1/1-sigma)-1/pa-pc(x)^1/1-sigma)*pa*(-1 + sigma))*2^3)/ (pa*(-3 + sigma)*(2 -
3*sigma + sigma^2)* (((1 + ((x^1/1-sigma)-1/pa-pc(x)^1/1-sigma)*pa)*sigma*2)/(-1 +
sigma))^sigma) + (A*(pa + 2)^2*(pa*(-2 + sigma) + 2))/ (pa^2*(-3 + sigma)*(2 - 3*sigma
+ sigma^2)* ((sigma*(pa + 2))/(-1 + sigma))^sigma)+ (A*1^2)/2 +
(A*((x^1/1-sigma)-1/pa-pc(x)^1/1-sigma)*(1 +
((x^1/1-sigma)-1/pa-pc(x)^1/1-sigma)*pa*(-1 + sigma))*1^3)/ (pa*(-3 + sigma)*(2 -
3*sigma + sigma^2)* (((1 + ((x^1/1-sigma)-1/pa-pc(x)^1/1-sigma)*pa)*sigma*1)/(-1 +
sigma))^sigma)+(A*(pa + 1)^2*(pa*(-2 + sigma) + 1))/ (pa^2*(-3 + sigma)*(2 - 3*sigma +
sigma^2)* ((sigma*(pa + 1))/(-1 + sigma))^sigma);

Answer 1

[Walter Roberson]

In that code, you could get that error if pc is a variable instead of a function and x is not a positive integer. Did you want pc*x instead of pc(x) ?

I note in your code that in several places you have

x^1/1-sigma

or

p(x)^1/1-sigma

I think it is unlikely that you want that code to be like that. x^1 is the same as x, and dividing by 1 would give you the same value you had, so x^1/1-sigma would be the same as x - sigma but I suspect you want x^(1/(1-sigma)) . I recommend, by the way, that you store that (1/(1-sigma)) into a variable in an earlier line and use the variable name instead of repeating the expression.

[Darcy Love]

function [AusProfInt]= AusProfInt (pa,AUeX,pc,sigma,A,x)
%sigma= 3;
%A= 100;
w= x^1/1-sigma;
k= w-1/pa-pc*w;

%x= 0.7;

Ausdemand1= -(A*2^2)/2 + (A*k*(1 + k*pa*(-1 + sigma))*2^3)/ (pa*(-3 + sigma)*(2 - 3*sigma + sigma^2)* (((1 + k*pa)*sigma*2)/(-1 + sigma))^sigma) + (A*(pa + 2)^2*(pa*(-2 + sigma) + 2))/ (pa^2*(-3 + sigma)*(2 - 3*sigma + sigma^2)* ((sigma*(pa + 2))/(-1 + sigma))^sigma)+ (A*1^2)/2 + (A*k*(1 + k*pa*(-1 + sigma))*1^3)/ (pa*(-3 + sigma)*(2 - 3*sigma + sigma^2)* (((1 + k*pa)*sigma*1)/(-1 + sigma))^sigma)+(A*(pa + 1)^2*(pa*(-2 + sigma) + 1))/ (pa^2*(-3 + sigma)*(2 - 3*sigma + sigma^2)* ((sigma*(pa + 1))/(-1 + sigma))^sigma);

Ausdemand2= pa.*Ausdemand1;
sumtotalAus= sum(Ausdemand2);

AusProfInt= pa*sumtotalAus - sum((Ausdemand2*AUeX));
end

[Darcy Love]

Changed the script. However, when i run the function the answer comes out as "NaN". Any ideas? Thanks a lot by the way, really appreciate your time

[Walter Roberson]

Your w has the problem I described above, that your code

x^1/1-sigma

is the same as x-sigma . Be generous with your use of brackets!

[Darcy Love]

function [AusProfInt]= AusProfInt (pa,AUeX,pc,sigma,A,x)
%sigma= 3;
%A= 100;
w= x^(1/1-sigma);
k= (w)-1/pa-pc*(w);
%x= 0.7;

Ausdemand1= -(A*2^2)/2 + (A*k*(1 + k*pa*(-1 + sigma))*2^3)/ (pa*(-3 + sigma)*(2 - 3*sigma + sigma^2)* (((1 + k*pa)*sigma*2)/(-1 + sigma))^sigma) + (A*(pa + 2)^2*(pa*(-2 + sigma) + 2))/ (pa^2*(-3 + sigma)*(2 - 3*sigma + sigma^2)* ((sigma*(pa + 2))/(-1 + sigma))^sigma)+ (A*1^2)/2 + (A*k*(1 + k*pa*(-1 + sigma))*1^3)/ (pa*(-3 + sigma)*(2 - 3*sigma + sigma^2)* (((1 + k*pa)*sigma*1)/(-1 + sigma))^sigma)+(A*(pa + 1)^2*(pa*(-2 + sigma) + 1))/ (pa^2*(-3 + sigma)*(2 - 3*sigma + sigma^2)* ((sigma*(pa + 1))/(-1 + sigma))^sigma);

Ausdemand2= pa.*Ausdemand1;
sumtotalAus= sum(Ausdemand2);

AusProfInt= pa*sumtotalAus - sum((Ausdemand2*AUeX));
end

[Walter Roberson]

1/1 is 1, so 1/1-sigma is the same as 1 - sigma. I suspect you want x^(1/(1-sigma))

Answer 2

[Darcy Love]

is this what you mean? I understand that i need to add brackets in. Even when i add this element in, it still comes out with "NaN"as the answer when i run the function.

Kind regards,