tag:blogger.com,1999:blog-3813701770708442620.post440335324830528283..comments2023-03-10T10:30:56.622-08:00Comments on G-FEED: Daily weather data: original vs knock-off solhttp://www.blogger.com/profile/00936469103707728475noreply@blogger.comBlogger1125tag:blogger.com,1999:blog-3813701770708442620.post-3319152706119544242014-04-13T06:52:05.422-07:002014-04-13T06:52:05.422-07:00Can you resolve some confusion on my part with the...Can you resolve some confusion on my part with the numbers in your table?<br /><br />You quote % reduction in out-of-sample RMSE from the various models. Take Corn model 1(c) for example. If it says that 1-sqrt(MSE)/sqrt(MSE0) = 0.2956 where MSE is the mean square cross-validation error, and MSE0 is the sample variance, then that implies prediction skill 1-MSE/MSE0 = 0.5038. This number is higher than R-squared (0.4519) which is clearly not plausible.<br /><br />Thanks, JoeJoehttp://www.joewheatley.netnoreply@blogger.com