I consistently get MC analysis results that are different from the quick calculation results (using the ecoinvent database and ReCiPe 2013, mid-point, hierarchy, climate change+fr. eutrophication+land use). I suspect that variation in 'distant' background processes is the cause. I wish to include variation from one of (my) background processes, since it has parameter values with variation that is central to my analysis. I do not want to include variation in, e.g., markets for energy or feed products, that are processes that I simply pick out of the ecoinvent db. Even if these processes are important for my result value, I want to exclude variation in their values, since it 1) is a possible cause for the diff. btw. MC and quick calc. results and, therefore, appear to represent an error somewhere, and 2) I want to calculate the input/output coefficient of variation for specific parameters, hence, variation of the output should only come from the input parameter variation that I focus on.
So my question regards how I can study and control the variation included in my model?
