hi, please I'm trying to guess a probability distribution(creating a null hypothesis) and create two datasets with approximately the same number of elements as the data I already have in my two data files , any idea on how this can be achieved?

hi, please I'm trying to guess a probability distribution(creating a null hypothesis) and create two datasets with approximately the same number of elements as the data I already have in my two data files , any idea on how this can be achieved?

[Walter Roberson]

Do you mean that you want to go through a list of distributions and randomly generate a number of points from each one and do kstest2 comparing the random data to the hypothesized distribution to check to see if that distribution "explains" the data?

[Precious Benmaurice]

Yes please

[John D'Errico]

Don't let the fact that your idea for how to find the best distribution is a bad one dissuade you. OK, I take that back. Do let it dissuade you, because it is a bad approach. Use the fitting tools in the stats toolbox. The link that Walter gave is a good start.

Answer 1

[Walter Roberson]

https://blogs.mathworks.com/pick/2012/02/10/finding-the-best/