72 Posts in 31 Topics by 15 membersJump to:
|Page: 1||Go to End|
|Author||Topic:Model Prediction||1307 Views|
27 May 2011 at 1:15pm
I have read in literature that to check the model prediction, the use of simple linear regression between the model predictiion and the actual values are of less usefull. What is your opinion?
28 June 2011 at 5:00pm
It depends on what you are wanting to find out. Generally, I am most interested in relative comparison between modeled and measured values, so a simple scatter plot with a R2 value will do. However, sometimes it's necessary to show statistical significance, in which case this approach is not useful.
|Go to Top|