tag:blogger.com,1999:blog-4630467186961819522.post1008003249219980451..comments2023-07-28T01:12:26.064-07:00Comments on Nierenberg's Climate Musings: Follow up on Rahmstorf 2007Nicolas Nierenberghttp://www.blogger.com/profile/15853996586598424891noreply@blogger.comBlogger4125tag:blogger.com,1999:blog-4630467186961819522.post-61404946774462610192009-12-11T20:21:39.245-08:002009-12-11T20:21:39.245-08:00I have included a plot of the actual versus predic...I have included a plot of the actual versus predicted values, as well as a plot of the residuals.<br /><br />I didn't understand your point about the DW statistic. Perhaps you could cite a reference? But I did include a plot of the residuals above, and it is obvious that they are highly autocorrelated.Nicolas Nierenberghttps://www.blogger.com/profile/15853996586598424891noreply@blogger.comtag:blogger.com,1999:blog-4630467186961819522.post-53490803542240181522009-12-11T16:37:36.190-08:002009-12-11T16:37:36.190-08:00Sorry to double post but, having had a look at Dur...Sorry to double post but, having had a look at Durbin-Watson tests I have a further question - D-W falls down when there's no constant term and I can't work out how either predicted or observed values can be constant, have I missed something? <br /><br />ThanksAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-4630467186961819522.post-88013524141298558122009-12-11T16:19:54.457-08:002009-12-11T16:19:54.457-08:00"First I plotted predicted values against the..."First I plotted predicted values against the actual values, and in fact they are not symmetrically distributed around either the diagonal or horizontal line. You can try it yourself from the code I already posted."<br /><br />Hi, I'm having trouble visualising what you are saying here and have neither the tech nor the chops to try it myself. Any chance you could post the plot?<br /><br />ThanksAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-4630467186961819522.post-36857937173257663412009-11-26T15:11:16.561-08:002009-11-26T15:11:16.561-08:00OT: I've replied to you at: http://scienceblog...OT: I've replied to you at: http://scienceblogs.com/stoat/2009/11/those_cru_emails_in_full.php#comment-2103576William M. Connolleyhttps://www.blogger.com/profile/05836299130680534926noreply@blogger.com