I am running PLP. How do we interpret the confidence band in the sensitivity graph? I assume the line graph represents the mean. the colored areas represent the area of the confidence boundaries. But in the 50% confidence boundary, the average line is outside the boundary. How can this be?
Thanks in advance for you help.
v/r,
john
Sensitivity Graphs and Confedence Intervals
-
- Member
- Posts: 47
- Joined: Mon Mar 14, 2016 4:58 pm
- Vensim version: PLE+
Sensitivity Graphs and Confedence Intervals
- Attachments
-
- Average is outside the 50% Confidence interval
- 50% Confidence.JPG (36.37 KiB) Viewed 1185 times
-
- Member
- Posts: 47
- Joined: Mon Mar 14, 2016 4:58 pm
- Vensim version: PLE+
Re: Sensitivity Graphs and Confedence Intervals
Maybe I am wrong about the line graph. Maybe the line is not the average of the simulation runs at each point?
Re: Sensitivity Graphs and Confedence Intervals
The line in this case is your base run for Test A. That suggests that the parameters of the sensitivity analysis are not centered around the base case.
I guess the reason we only see the 50th percentile here is that the 75/95/100 values are set to white?
I guess the reason we only see the 50th percentile here is that the 75/95/100 values are set to white?
/*
Advice to posters (it really helps us to help you)
http://www.ventanasystems.co.uk/forum/v ... f=2&t=4391
Blog: http://blog.metasd.com
Model library: http://models.metasd.com
Bookmarks: http://delicious.com/tomfid/SystemDynamics
*/
Advice to posters (it really helps us to help you)
http://www.ventanasystems.co.uk/forum/v ... f=2&t=4391
Blog: http://blog.metasd.com
Model library: http://models.metasd.com
Bookmarks: http://delicious.com/tomfid/SystemDynamics
*/