I have some non-parametric data and I would like to determine the expected value. I can't find any guidance on this in my old stat books (I find formulas for the EV of various distributions, but nothing non-parametric).
Background: I ran a Monte Carlo simulation on a model for a business decision. The simulation ran 329,000 iterations. I'm trying to determine the expected value of the output. The data is heavily left-skewed. I ran a distribution identification test in Minitab and no distribution was found to have a good fit (I assume because of the large sample size). Not sure where to go from here.
I need to identify the expected value to use for comparisons between different options, as well as for calculating the value of information.
Here is a screengrab from Minitab of the graphical summary, with P-P plots from Minitab distribution identification tests below.
