Store value into matrix, and then perform calculation upon each element

Hi, I have a question about performing calculation on each matrix elements.

My code is here:

syms m a x u;

a=16807;

m=2^31-1;

x=1;

for i=1:10^2

    y=a*x

    x=mod(y,m)

    u(i)=x/m

    X = betainv(u,2,3);  % beta distributed random fraction of defective parts

    Y = logninv(u,5,3); % lognormal distributed demand amount

    Q = 100;

    Revenue=min(Y,(1-X)*Q)*53;  %the total revenue

    Cost=27*Q;

    Salvage=3*max(Q*(1-X)-Y,0)+X*Q;   %all the unsolde parts are salvaged, defective parts included

    p=Revenue+Salvage-Cost;

end;

expectedprofit=mean(p)

Basically, I generate a series of uniformly distributed random numbers. And then I use these numbers to genearate beta and lognormal distributed ones, and I would like to store them into matrix X and Y. Then I want to perform the profit calculation (revenue-cost+salvage cost etc.) for each of the values in X and Y. I just don't know how to do that.

I tried to put things as X(i) and Y(i) (to store value into matrix), but MLATLAB finds an error in that step.

I just started learning Monte Carlo simulation using MATLAB. Hope somebody can help me out here.

Thanks a lot! Ying

Answer 1

[Ondrej]

I guess you want to do

    X(i) = betainv(u(i),2,3);
    Y(i) = logninv(u(i),5,3);

you either missed the u(i) index, or you want to store growin u, in that case X(:,i), Y(:,i) would be some matrices (but with growing columnsize..size in your case u is growing).

[Ying]

Hi Ondrej,

Thanks for such quick answer.

I tried your way of writing X(:,i) and Y(:,i) and it works!

But the other way your suggested which is to write up u(i) and X(i) and Y(i) still yields error message in MATLAB.

The u value is growing from 1 column to 100 column as I put in the loop,and I am puzzled that why wouldn't

X(i) = betainv(u(i),2,3);

Y(i) = logninv(u(i),5,3);

work? What's different from writing X(:,i) and Y(:,i)?

Thanks for helping me out!

Ying

[Ondrej]

Well, the difference is that X(i) means X is a vector and X(:,i) is a matrix (meaning the i-th column of X). See http://www.mathworks.com/help/techdoc/math/f1-85462.html

Btw. if you do for loops with variables growing inside, consider preallocating: http://www.mathworks.com/support/solutions/en/data/1-18150/

As for your code, it doesn't work for me because of the symbolic variables in betainv, so I think I can't help you with that.