where can i use round function to make the function perfect

function [amt_ret, x]= calculate_change(tendered,price)
%--------------------------------------------------------------------------------------------------------%
%     A. peter                                                           2-24-2018                               %
% This function calculates the change from a purchase given the                        %
% amount tendered and the purchase price. The function also provides            %
% minimum denominations to make up the change.                                             %
%                                                                                                                                  %
% Calling syntax:                                                                                                        %
%    [amt_ret, xprice] = calculate_change(tendered,price)                                    %
%                                                                                                                                  %
%  inputs: tendered = amount of currency given to the cashier [$]                        %
%          price    = purcahse price [$]                                                                         %
%                                                                                                                                  %
%   output:   change = tendered - price [$]                                                               %
%                           x  = vector indicating the amountof each                                   %
%                                  denomination to make up the change                                 %
%                                                                                                                                   %
%---------------------------------------------------------------------------------------------------------%
%
format bank
denoms=[50,20,10,5,1,0.25,0.10,0.05,0.01];    % denominations vector
change=tendered-price;
for k=1:length(denoms)
    index=0;
    while change+75*eps>=denoms(k)
    index=index+1;
    change=change-denoms(k);
    end
x(k)=index;
end
amt_ret=sum(denoms.*x);
error = change - amt_ret;
end

[Image Analyst]

What's your definition of "perfect"?

http://www.mathworks.com/matlabcentral/answers/13205#answer_18099

Answer 1

[Roger Stafford]

I would suggest initially multiplying both 'tendered' and 'price' by 100 and using 'round' on the result:

tendered = round(100*tendered);
price = round(100*price);

so as to represent pennies instead of dollars. Similarly your 'denoms' should each be 100 times as great. After that you will be working strictly with integers and have no need for such stuff as "while change+75*eps>=denoms(k)". In fact you can use the 'mod' function to accomplish in one step a whole 'while' loop computation. Of course at the last step you need to divide 'amt_ret' by 100 to get back to dollars.

[Roger Stafford]

Instead of 'mod' it is easier to use 'floor' and division by 'denoms'. Since only integers are involved there should be no errors.

change = tendered-price;
denoms = [5000,2000,1000,500,100,25,10,5,1];
m = length(denoms);
x = zeros(1,m);
for k = 1:m
  x(k) = floor(change/denoms(k));
  change = change-x(k)*denoms(k);
end
amt_ret = sum(denoms.*x)/100;