I would like to know the explicit structure of the built-in Vensim function "RANDOM PINK NOISE" I am aware that it must be some form of stock-adjustment structure to build in the autocorrelation. But it seems to be different from the versions that Sterman has in its appendix to his textbook.
I have read the help information on the function, but could not quite make sense of it. It talks about a version by Ed Anderson and also about an article by Richardson and Pugh 1981. It might be helpful for me to know which two articles are meant here.
Explicit Stock-Flow Structure of the Pink Noise
-
- Super Administrator
- Posts: 4599
- Joined: Wed Mar 05, 2003 3:10 am
Re: Explicit Stock-Flow Structure of the Pink Noise
It comes from Tom's blog, but I think there was a slight modification before we put it directly into Vensim.
https://metasd.com/2010/03/pink-noise/
https://metasd.com/2010/03/pink-noise/
Code: Select all
:MACRO: RANDOM PINK NOISE(mean, std deviation, correlation time, seed)
RANDOM PINK NOISE = INTEG(updating pink noise,mean+std deviation*RANDOM NORMAL(-6,6,0,1,seed))
~ mean
~ |
updating pink noise = (scaled white noise-RANDOM PINK NOISE)/correlation time
~ mean/correlation time
~
|
scaled white noise =mean+white noise*std deviation*SQRT((2-time step$/correlation time)/(time step$/correlation time))
~ mean
~ |
white noise = RANDOM NORMAL(-6,6,0,1,seed)
~ dmnl
~ |
:END OF MACRO:
Advice to posters seeking help (it really helps us to help you)
http://www.ventanasystems.co.uk/forum/v ... f=2&t=4391
Units are important!
http://www.bbc.co.uk/news/magazine-27509559
http://www.ventanasystems.co.uk/forum/v ... f=2&t=4391
Units are important!
http://www.bbc.co.uk/news/magazine-27509559