forum.alglib.net http://forum.alglib.net/ |
|
quest about least squares http://forum.alglib.net/viewtopic.php?f=2&t=627 |
Page 1 of 1 |
Author: | tianxinzhe [ Fri Oct 12, 2012 3:04 am ] |
Post subject: | quest about least squares |
hellp everyone,I have data below x , y 2001,75330 2002,75636 2003,71422 2004,70952 2005,75398 2007,78281 2008,79524 2009,72624 I want to fitting the Y ,when x equal 2006. which method can be chosen. |
Author: | Sergey.Bochkanov [ Fri Oct 12, 2012 5:42 am ] |
Post subject: | Re: quest about least squares |
With such small amount of data and such large non-monotonic variations it is better to use something simple. Complex methods won't work. I think, it is better to calculate Y(2006) as 0.5*(Y(2005)+Y(2007)). You may think that it is too simple, but the problem is that there is no magic formula which can predict sudden changes like ones which happened in 2003 (75000 to 71000) and 2005 (70000 to 75000). |
Page 1 of 1 | All times are UTC |
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |