monte carlo random number generation
18 views (last 30 days)
stavros boutsikos on 29 Mar 2021
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?
Andreas Apostolatos on 29 Mar 2021
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,
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,
I hope that this information helps.
More Answers (1)
Image Analyst on 29 Mar 2021
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.