/sci/ - Science & Math

Hi, sorry for asking such a stupid question about probability distributions, but I don't know where else to ask it. I have a set of numbers, all of them very small and all positive (between 0 and 1). I can take the standard deviation of this set and its mean, and I would like to create some number that "fits in" with this set, however if I put the standard deviation and mean into a gaussian distribution I end up with a lot of negative numbers (I can obviously "clip" this by just taking the absolute value, but this leaves the distribution biased towards the values between the mean and 0). Is there some other form of probability distribution I should be using here? Again, sorry for the basic question, I'm here from /g/ just trying to implement a method for accurately guessing at missing data.