Setting a probability from e.g., 0% to 4% results in some negative numbers in the sample.
agdfoster opened this issue · comments
If I set 0% to 4% as a probability input, with a normal distribution some of my samples will be negative, which can cause lots of impossible outcomes further down the chain.
I think this is basically just the software not intuiting that most of the time 0% is the absolute lowest number and the curve shouldn't be allowed to go beyond it.
Not sure if a quick fix would be to set 0.00 as the lowest value sample if a user inputs a percentage.
E.g., if user types 4% rather than 0.04, guesstimate interprets that as 0.04 but also sets the samples to have a minimum value of 0.00.
(current work around is to set a higher value lowest probability, but this is skewing my results.
+1 for a function that makes negative values impossible
You can use =filterLessThan(WAR,0) to filter all results less than 0. It's one of the few features not added to the documentation yet.
filterLessThan,
filterGreaterThan,
filterBetween
all should work