date: Tue, 15 Jul 2003 09:11:30 -0400
from: Edward Cook <drdendroatXYZxyzo.columbia.edu>
subject: Re: Fwd: Climatologists have used outdated time series analysis
to: Keith Briffa <k.briffaatXYZxyz.ac.uk>
Sounds good. By the way, in what sense is the Esper domain too small.
I freely admit that it is best suited for the extra-tropics
(ca. >30N) and tried to make that painfully obvious in the Esper et
al. paper. So a critique of it based on domain size is self-serving,
as if it was not pointed out in the Esper et al. paper. There is no
way that you can say that the Mann et al. paleo-data domain to proper
for a NH reconstruction either, particularly back before ca. 1600,
since it is mostly based on >30N data as well. I would expect that
issue to be mentioned as well in the Rutherford paper, but I am sure
it was not for reasons that we both understand. I am afraid that the
playing field is not fair in general.
Cest le vie, re Italy.
>Will get a copy of this and send. Of course I agree 100 per cent re
>"optimal methods" (yuk) versus traditional (dare I say , even local
>point ) regressions - in fact I am actually an author on a Scott
>Rutherford (and others including Mann) paper that shows just this.
>Incidentally , this also concludes that differences between Mann and
>Esper NH curves are mostly a matter of spatial domain difference
>(with yours too small of course ) .
>Be careful not to oil yourself too much because the smell of
>grilling fat will annoy the neighbours. Seems like the timing of
>your Italy jaunt does not suit us by the way so I think you are safe
>as regards a visit.
>At 06:20 AM 7/15/03 -0400, you wrote:
>>Outdated as of June 28, 2003? Guilty as charged I guess. I'm not
>>familiar with this paper nor the authors. Of course I am skeptical.
>>In comparing my old fashioned least squares methods with advanced
>>'optimal' methods like RegEM (that Mike is enamored with) and
>>hierarchical Bayes, there is fuck-all difference in the results.
>>Connie Woodhouse's results with neural networks doesn't show much
>>either over linear regression. If you are able to get a pdf, please
>>email it to me. I am not in position to get it now. Am at the beach.
>>>>Date: Mon, 14 Jul 2003 17:01:56 +0100 (BST)
>>>>From: Timo Hameranta <timohameatXYZxyzoo.co.uk>
>>>>Subject: Climatologists have used outdated time series analysis
>>>>To: firstname.lastname@example.org, email@example.com, firstname.lastname@example.org,
>>>> email@example.com, firstname.lastname@example.org, email@example.com,
>>>> firstname.lastname@example.org, email@example.com, firstname.lastname@example.org,
>>>> email@example.com, wsoonatXYZxyz.harvard.edu
>>>>Dear all, see the study
>>>>Godtliebsen, F., L. R. Olsen, and J.-G. Winther, 2003.
>>>>Recent developments in statistical time series
>>>>analysis: Examples of use in climate research,
>>>>Geophys. Res. Lett., 30(12), 1654,
>>>>doi:10.1029/2003GL017229, June 28, 2003.
>>>>In this paper we present some recently developed time
>>>>series analysis methods. Further, we apply these
>>>>methods to a suite of climatological and synthetic
>>>>time series. We show what information (or statistical
>>>>significance) that can be drawn from such time series
>>>>and which otherwise, i.e. by simpler methods, would be
>>>>difficult to extract. We conclude by recommending the
>>>>use of advanced statistical time series analysis for a
>>>>wide range of applications connected to studies of
>>>>climate variability and climate change.
>>>>Yahoo! Plus - For a better Internet experience
>>>Professor Keith Briffa,
>>>Climatic Research Unit
>>>University of East Anglia
>>>Norwich, NR4 7TJ, U.K.
>>Dr. Edward R. Cook
>>Doherty Senior Scholar and
>>Director, Tree-Ring Laboratory
>>Lamont-Doherty Earth Observatory
>>Palisades, New York 10964 USA
>Professor Keith Briffa,
>Climatic Research Unit
>University of East Anglia
>Norwich, NR4 7TJ, U.K.
Dr. Edward R. Cook
Doherty Senior Scholar and
Director, Tree-Ring Laboratory
Lamont-Doherty Earth Observatory
Palisades, New York 10964 USA