cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

Community Tip - When posting, your subject should be specific and summarize your question. Here are some additional tips on asking a great question. X

Determine Relative Contribution of Sources of Variance | Monte Carlo

Theo_Gagner
6-Contributor

Determine Relative Contribution of Sources of Variance | Monte Carlo

Hello,

 

I have a working simulation worksheet that runs a Monte Carlo analysis on a function.  With the results, I can bin the simulations for the input variables as well as for the output.  I am struggling though, to find references on how I might "tease" out the relative contribution of sources of variance.  

 

Are there any suggestions for using the input function, along with known means and standard deviations of the input variables to establish these values?  

 

Is there an analytical way to look at the result of the Monte Carlo simulations and determine relative variance?

 

Sorry for the abstraction in advance...I attached my Monte Carlo simulation up to this point, but couldn't even begin to try and "mock up" what I am asking for help with.

1 ACCEPTED SOLUTION

Accepted Solutions
LucMeekes
23-Emerald III
(To:Theo_Gagner)

A little experimenting:

LM_20181128_Variance.png

That's pretty close to the result of your MonteCarlo run's stdev (4300 N)...But no proof!

 

Success!
Luc

View solution in original post

3 REPLIES 3

I can't open your file (5.0?).  But here's a sheet I've posted before that makes a tornado graph to compare contributions.

LucMeekes
23-Emerald III
(To:Theo_Gagner)

You might want to look here

https://en.wikipedia.org/wiki/Variance#Product_of_independent_variables

for information on symbolically/analytically determining the variance of your function result from the variances (and expected values) of its input variables.

 

Success!

Luc

LucMeekes
23-Emerald III
(To:Theo_Gagner)

A little experimenting:

LM_20181128_Variance.png

That's pretty close to the result of your MonteCarlo run's stdev (4300 N)...But no proof!

 

Success!
Luc

Top Tags