monte carlo random number generation

i want to use a function that generates a random number from [1,2,3,4,5,6,7,8,9,10] but the probabilities of each number are different. For exaample the pof drawing one is 0.2 but p of drawing 3 is 0.06? is there a function or a code that i can use in order to generate a random number by taking into consideration that each number has different weighting?

Answer 1

[Andreas Apostolatos]

Hello,

It seems that what you are seeking for can be utilized using function 'randsample' when using the variant that takes as argument variable 'population', which in your case is vector '[1,2,3,4,5,6,7,8,9,10]', and an additional vector of weights 'w' that has the same size as variable 'population', see the following documentation link for more information,

https://www.mathworks.com/help/stats/randsample.html#d123e694964

Note that function 'randsample' is using a uniform probability distribution and it is provided as part of the Statistics and Machine Learning Toolbox. The topic is also discussed in the following MATLAB Answers thread where additional workflows are discussed,

https://www.mathworks.com/matlabcentral/answers/502828-generating-a-weighted-random-number

I hope that this information helps.

Kind Regards,

Andreas

[stavros boutsikos]

thank you very much Andrea, it was really helpfull and it helped me realise that my code was working and i was just overthinking it.

Answer 2

[Image Analyst]

You can simply get the cumulative sum and compare your random number to that. For example

r = rand
c = cumsum(p)
output = find(r < c, 1, 'last')

If that doesn't work for you, then give us all the probability values.

[stavros boutsikos]

thank you very much, it actually helped me understand that the code i had was actually working!