tag:blogger.com,1999:blog-28837843.post2628412422809555849..comments2024-03-19T07:10:27.303-07:00Comments on Quark Soup by David Appell: Entertaining Roy Spencer's Fit to His Temperature DataDavid Appellhttp://www.blogger.com/profile/03318269033139447591noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-28837843.post-84781040007540461672012-01-10T03:49:56.761-08:002012-01-10T03:49:56.761-08:00It is not quite the thing I was thinking of, but h...It is not quite the thing I was thinking of, but http://tamino.wordpress.com/2012/01/09/steps/ discusses the same issue: that more params leads to a better fit, but may not be a better model. See AIC, there.William M. Connolleyhttps://www.blogger.com/profile/05836299130680534926noreply@blogger.comtag:blogger.com,1999:blog-28837843.post-76574826389152163202012-01-07T17:28:33.346-08:002012-01-07T17:28:33.346-08:00funny. i always thought it would be more "ent...funny. i always thought it would be more "entertaining" to use a linear fitbobnoreply@blogger.comtag:blogger.com,1999:blog-28837843.post-87640572755984704792012-01-04T03:35:33.400-08:002012-01-04T03:35:33.400-08:00> a 6th-order fit has a better correlation coef...> a 6th-order fit has a better correlation coefficient than does the 3rd-order fit<br /><br />But you expect that anyway; you're allowing more parameters, so you get a better fit.<br /><br />There is a proper statistic-y way of accounting for this, but I forget it. DC will know :-)William M. Connolleyhttps://www.blogger.com/profile/05836299130680534926noreply@blogger.com