Comments on: Confidence Limits of Minimum Roughness Criterion http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/ The Power of Numeracy Wed, 16 May 2012 18:37:00 +0000 hourly 1 http://wordpress.org/?v=3.3.2 By: Niche Modeling » Examples of simple smoothers http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-4760 Niche Modeling » Examples of simple smoothers Fri, 18 Apr 2008 01:57:52 +0000 http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-4760 [...] How much error is there in smoothed climatic and financial series? How much does variability at the ends of the series affect the trend as a result? Here we showed that certain ways of treating the end points introduce a lot of variability. Here we show in certain smoothers variability at the ends can affect the whole smooth! [...] [...] How much error is there in smoothed climatic and financial series? How much does variability at the ends of the series affect the trend as a result? Here we showed that certain ways of treating the end points introduce a lot of variability. Here we show in certain smoothers variability at the ends can affect the whole smooth! [...]

]]> By: UC http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-4759 UC Sat, 12 Apr 2008 07:23:49 +0000 http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-4759 This topic is in my list 'pictures you won't see at RC' ( http://signals.auditblogs.com/2008/01/03/some-interesting-figures/ ) Indeed, these padding procedures are very sensitive to measurement noise. And the effect near the end-points can be quite easily computed with error propagation formulas (see for example http://www.climateaudit.org/?p=2541#comment-188580 ) . But in these climate papers signal, process noise , measurement noise are not defined well, so they can use any method they prefer. This topic is in my list ‘pictures you won’t see at RC’ ( http://signals.auditblogs.com/2008/01/03/some-interesting-figures/ )

Indeed, these padding procedures are very sensitive to measurement noise. And the effect near the end-points can be quite easily computed with error propagation formulas (see for example http://www.climateaudit.org/?p=2541#comment-188580 ) . But in these climate papers signal, process noise , measurement noise are not defined well, so they can use any method they prefer.

]]> By: UC http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-6368 UC Sat, 12 Apr 2008 07:23:00 +0000 http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-6368 This topic is in my list 'pictures you won't see at RC' ( http://signals.auditblogs.com/2008/01/03/some-interesting-figures/ ) Indeed, these padding procedures are very sensitive to measurement noise. And the effect near the end-points can be quite easily computed with error propagation formulas (see for example http://www.climateaudit.org/?p=2541#comment-188580 ) . But in these climate papers signal, process noise , measurement noise are not defined well, so they can use any method they prefer. This topic is in my list ‘pictures you won’t see at RC’ ( http://signals.auditblogs.com/2008/01/03/some-interesting-figures/ )

Indeed, these padding procedures are very sensitive to measurement noise. And the effect near the end-points can be quite easily computed with error propagation formulas (see for example http://www.climateaudit.org/?p=2541#comment-188580 ) . But in these climate papers signal, process noise , measurement noise are not defined well, so they can use any method they prefer.

]]> By: admin http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-4758 admin Fri, 11 Apr 2008 01:59:18 +0000 http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-4758 That is right. What is frustrating to me is that this is the sort of authoritative paper that convinces world governments we are all going to warming hell in a handbasket. That is right. What is frustrating to me is that this is the sort of authoritative paper that convinces world governments we are all going to warming hell in a handbasket.

]]> By: admin http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-6367 admin Fri, 11 Apr 2008 01:59:00 +0000 http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-6367 That is right. What is frustrating to me is that this is the sort of authoritative paper that convinces world governments we are all going to warming hell in a handbasket. That is right. What is frustrating to me is that this is the sort of authoritative paper that convinces world governments we are all going to warming hell in a handbasket.

]]> By: lucia http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-4757 lucia Thu, 10 Apr 2008 22:34:24 +0000 http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-4757 Well, it had struck me that the uncertainty in correct location of the smooth line had to be high, and particularly near the end point. So, it would appear that if Ramhstorf7 had given the slightest consideration to uncertainty, they would have seen that they had no clue where the non-linear fit went through the data! Well, it had struck me that the uncertainty in correct location of the smooth line had to be high, and particularly near the end point. So, it would appear that if Ramhstorf7 had given the slightest consideration to uncertainty, they would have seen that they had no clue where the non-linear fit went through the data!

]]> By: lucia http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-6366 lucia Thu, 10 Apr 2008 22:34:00 +0000 http://landshape.org/enm/confidence-limits-of-minimum-roughness-criterion/#comment-6366 Well, it had struck me that the uncertainty in correct location of the smooth line had to be high, and particularly near the end point. So, it would appear that if Ramhstorf7 had given the slightest consideration to uncertainty, they would have seen that they had no clue where the non-linear fit went through the data! Well, it had struck me that the uncertainty in correct location of the smooth line had to be high, and particularly near the end point. So, it would appear that if Ramhstorf7 had given the slightest consideration to uncertainty, they would have seen that they had no clue where the non-linear fit went through the data!

]]>