how to repeat a loop until a condition is met than draw histogramm

1 view (last 30 days)
what i want is to generate 2 uniforms randoms numbers 'u1' and 'u2' and evaluate the magnetude of the 2 nummbers .if the magnetude "d<1" it is transformed and scalled by u1 to finally draw a histogramm of the outpout result.
the pgroramm is looks like this .
repeat
u1=2*rand()-1;
u2=2*rand()-1;
d=(u1.^2)+(u2.^2);
until 0<d<1
G=sqrt((-2*log(d))/d);
return G*u1

Accepted Answer

Yongjian Feng
Yongjian Feng on 31 Aug 2021
Edited: Yongjian Feng on 31 Aug 2021
Just use a while loop?
Try:
done = false;
while ~done
u1=2*rand()-1;
u2=2*rand()-1;
d=(u1.^2)+(u2.^2);
done = d > 0 && d < 1;
end
d
G=sqrt((-2*log(d))/d);
G*u1
  1 Comment
ad lyn
ad lyn on 31 Aug 2021
and what's about the histogram? i need to draw the histogram of (G*u1) so i can get a gaussian distribution

Sign in to comment.

More Answers (1)

Image Analyst
Image Analyst on 31 Aug 2021
You're not going to be able to generate many points before you get a squared distance of less than 1 and need to quit. I ran it several times and it was always done in the first pass - just was able to get one point. Nonetheless, you need to index your distance so you have an array of distances that you can pass into the histogram() function. Here is how you'd do it:
loopCounter = 1;
maxIterations = 100000; % Failsafe
% Preallocate space
distanceSquared = zeros(1, maxIterations);
while loopCounter <= maxIterations
u1=2*rand()-1;
u2=2*rand()-1;
distanceSquared(loopCounter) = (u1.^2)+(u2.^2);
% See if we need to quit. Quit when distanceSquared is less than 1.
if distanceSquared(loopCounter) < 1
% Crop off unused.
distanceSquared = distanceSquared(1 : loopCounter);
break;
end
loopCounter = loopCounter + 1;
end
caption = sprintf('Was able to generate %d distances.\n', loopCounter);
fprintf('%s\n', caption);
% Compute G from distanceSquared:
G = sqrt((-2*log(distanceSquared)) ./ distanceSquared);
histogram(G);
grid on;
title(caption);
xlabel('G');
ylabel('Count');

Community Treasure Hunt

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

Start Hunting!