47 views (last 30 days)

clc; clear all; close all;

n = 200; % Number of iteration

ro = 0.75; % Correlation factor

X = zeros(1,n);

x = zeros(1,n);

for ii = 1:n

x = rand(1,ii);

y = rand(1,ii);

X = x*ro+y*sqrt(1-ro^2);

if X > 1 % If X > 1 start iteration no. ii again until X is below 1

ii = ii - 1;

end

end

plot(x,X,'+')

Niels
on 25 Feb 2015

Edited: Niels
on 25 Feb 2015

Change your for-loop to a while-loop if you want to keep looping. I also believe that you stored your data incorrectly, as x, y and X were not stored at every time step, while your random number generation grows larger for every iteration ii...

clc; clear all; close all;

n = 200; % Number of iteration

ro = 0.75; % Correlation factor

X = zeros(1,n);

x = zeros(1,n);

y = zeros(1,n);

ii = 1;

while (ii <= n)

x(ii) = rand();

y(ii) = rand();

X(ii) = x(ii)*ro+y(ii)*sqrt(1-ro^2);

if (X(ii) <= 1) % Don't increment if X > 1.

ii = ii + 1;

end

end

figure, plot(x,X,'+')

Joseph Cheng
on 24 Feb 2015

you'd want to use "continue" to skip the iteration.

for ind =1:5

if ind==3

continue

else

disp(ind)

end

Sean de Wolski
on 24 Feb 2015

Why not just only generate random values that meet your criteria? Here's an example:

% Solve for x

syms x ro y

X = x*ro+y*sqrt(1-ro^2);

xx = solve(X,x)

% Plug in values

yval = -0.1;

roval = 0.3;

xmax = double(subs(xx,{y, ro},{yval,roval}));

% Generate only random numbers that meet criteria

xrand = rand(100,1)./xmax;

% Create X

Xfinal = xrand.*roval+yval*sqrt(1-roval^2);

% Check all X < 1

assert(all(Xfinal < 1))

Roger Stafford
on 25 Feb 2015

Edited: Sean de Wolski
on 25 Feb 2015

I am guessing that what you actually want is the following. It will give you 200 values in X which are all less than or equal to 1.

X = zeros(1,n);

x = zeros(1,n);

for ii = 1:n

b = true;

while b

t1 = rand;

t2 = t1*ro+rand*sqrt(1-ro^2);

b = t2 > 1;

end

X(ii) = t2;

x(ii) = t1;

end

plot(x,X,'+')

Opportunities for recent engineering grads.

Apply TodayFind the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!
## 0 Comments

Sign in to comment.