Comment on: “Archibald:Where The Science Is.”

Advertisements

37 Responses to “Comment on: “Archibald:Where The Science Is.””

  1. JohnZ Says:

    Moving here so the turkey can’t wipe my posts. Let the record stand clear and proud!

    Now, I’ve already answered Graeme’s question twice (both wiped) so here it is again:

    I would use all available stations. If I distrusted the data from some stations, I would explain why, define some exclusion criteria and then use the remaining stations.

    Archy has excluded the urban stations, with a clear explanation. However, he also excluded the vast majority of rural stations and has not explained any reason for doing so.

    Now turkey, the only way you can rescue Archy from charges of cherry picking is by showing us the criteria he used to exclude all but 5 stations.

  2. graemebird Says:

    He is being cost-effective. And he is relating modern information to the earlier information he had from Europe two centuries before.

    Less is more. You pick 5 stations you can be sure of and it will give you an honest indication but the error bars will be higher. This effort to include all the available data is not cost-effective and you can always go back for more.

    If we look at the global averages they were not only buggered by the heat island effect. But they were also buggered by the collapse of the Soviets, leading to the Siberian stations closing down. Thus giving us a spurious higher average figure for the 90’s. By choosing 5 you save time and money and if more time and money become available you can then go back and pick out 25 that you can be sure of.

    The proof is in the pudding. Archibalds study came through with the goods and was robust enough to anticipate the corrections in the data that McIntyre forced on NASA.

    Archibald is an extremely solid scientist. And I would say this comes from working in industry as much as anything else.

  3. JohnZ Says:

    You pick 5 stations you can be sure of and it will give you an honest indication

    Lordy. How could he be “sure of” the 5 stations he selected? Is this stated anywhere in the paper?

    If you knew how to use these new fangled computer thingimy jiggs, you’d know it takes the same effort to analyse 500 stations as it does 5.

    This is the clearest case of cherry picking I have ever seen.

  4. graemebird Says:

    Well he was sure of them or he would not have picked them surely.

    What is your point here?

  5. graemebird Says:

    There was no case of cherry picking. You are lying.

  6. JohnZ Says:

    Well he was sure of them or he would not have picked them surely.

    Do you know what “circular logic” is, turkey? You assume a-priori that Archy is trustworthy, and so give him a pass on his cherry-picking methodology.

    Now prove those 5 stations were trustworthy or retract your nonsense.

  7. graemebird Says:

    The idea is to pick five that you are pretty sure are good. Thats quick and cost effective and if you think you need more you can go back for more. He’s been powerfully vindicated since he anticipated corrections to the general pool of data.

  8. JohnZ Says:

    The idea is to pick five that you are pretty sure are good.

    And how did he know they were good?

    I’m getting tired of this filibuster, turkey.

  9. graemebird Says:

    Well he obviously thought they were pretty good or he would not have picked them. No use picking 5 that are no good and skewing your result. He might have related them to the 5 he had in Europe. He may have judged them on the basis that he didn’t want changes in the gulf stream to mess them up. He may have done a bit of background to satisfy himself of the solid record-keeping practices and sound location of the five.

    But you want to pick 5 good ones. Lest you skew the results. And the more you pick the harder is quality control.

  10. JohnZ Says:

    Turkey turkey turkey…. WHY DID HE RESTRICT HIMSELF TO 5 STATIONS?

    HOW DID HE KNOW THEY WERE GOOD?

  11. Lyam Says:

    Nice effort JohnZ.

    We all knew Birdy was a circular logic master, but you managed to pinpoint it beautifully. He usually manages to deflect the question by wandering off topic bringing in all sorts of unrelated crap.

    Kudos to you.

  12. Jason Soon Says:

    the emperor has no clothes

    and it ain’t a pretty sight

    women don’t look or you’ll be converted to lesbianism …

  13. JohnZ Says:

    Are you ready to concede defeat turkey?

  14. graemebird Says:

    Why not 5 stations. It linked up with his Europe data.

    I’m mentioned this about ten times you cunt.

    I’ll mention it again. IT LINKED UP WITH HIS HISTORICAL EUROPEAN DATA.

    5 good stations is good. Your thesis that he needed more stations is wrong.

    That takes more time and therefore is superfluous to requirements.

    And the proof is there. His findings were superior to anything prior until such time as McIntyre forced NASA to update their data.

  15. JohnZ Says:

    Hilarious turkey, so if I choose 5 stations and they show a warming trend, is AGW proved since it links up with the worldwide IPCC data?

    If Archy wants to be taken seriously as a scientist rather than as a West Australian grape grower he’s going to have to put in the leg work. 5 stations is not sufficient by ANY STANDARD, let alone 5 stations MANUALLY SELECTED.

    This paper would not pass muster in a first-year undergraduate science class.

    You should listen to the science workers turkey, they know more than you.

  16. graemebird Says:

    Look I keep explaining and you keep pretending that I haven’t already explained. His study was superior and if he had needed to make it a bit better with a little bit more time and money he could have done that. Cost-effectiveness is important in the private sector. You might call it cheap and nasty but this is why science must be private. To get great results at low cost.

    Including more stations would have required greater quality control and it would have lead the one set of data to be out of kilter with the European stuff.

    I keep on explaining. I’m right and you just do not want to accept it.

  17. graemebird Says:

    You are a just a cunt mate. And you and Humphreys ought both be euthanised to save me the trouble of having to explain the same thing to you guys over again.

  18. JohnZ Says:

    Including more stations would have required greater quality control and it would have lead the one set of data to be out of kilter with the European stuff.

    GOTCHA!!!!

    You just admitted he massaged the data to mimic europe. Why didn’t you just say that to begin rather than filibustering?

  19. graemebird Says:

    No I didn’t you are lying. He didn’t massage anything.

    If he massaged the data he wouldn’t have gotten the right result. So there is no point in it. Thats what you leftists do.

  20. JohnZ Says:

    Including more stations would have required greater quality control and it would have lead the one set of data to be out of kilter with the European stuff.

    Explain to us turkey how increasing the sample side would make the data “out of kilter with europe”….

  21. JohnZ Says:

    side=size

  22. graemebird Says:

    Well from memory the data he had for the Europeans from a couple of centuries before only had about 5 stations. You may be able to correct me on this. So there would have been a tendency to match like with like unless you really needed to go back and spend more time and money and get a result with greater statistical significance.

    Good data is preferable to bad data with allegedly greater statistical significance. Archibalds less is more approach is what you would see in the public sector a lot.

    The taxeaters fucked up entirely when they didn’t cull down the data. So the whole world wound up thinking the 90’s were hotter than they really were. And partly simply because the Soviets lost their empire and their Siberian stations at the same time.

    By including more data then they ought to have they mislead. But Archibalds study didn’t mislead and could always have been checked and added to.

    Basically he was acting more like a good journalist then a scientist here and letting us know about the scoop. But even with the small amount of original work he did it was solid work.

    There is another thing that the private sector would do. No use reinventing the wheel. Archibalds summary of what was already known was excellent too and there was a small truncated study of his own on top of that.

    Did the job brilliantly.

  23. Mark Hill Says:

    Now Graeme stop being a cunt. We all know that there are extremists on either side of the science of the warming debate.

    A long, high frequency data series from a number of stations spread throughout with a high degree of data quality is needed. As an economist trying to make some sense out of it, the actions on some scientists on either side isn’t useful. A maniac, loud-mouth layman as yourself adding to the cacophony isn’t bloody helping.

    Z is actually being neutral here. He hasn’t taken an alarmist position but has sensibly noted that five weather stations is simply inadequate for adult analysis of a serious issue. Now stop acting like a child.

  24. graemebird Says:

    But it wasn’t needed if cost-effectiveness was what you were after. Since Archibald managed to get results which anticipated subsequent corrections to the data. Cost-effectiveness matters.

  25. Mark Hill Says:

    RIIIGHT.

    You expect us to believe us that you’d be happy with a similar response from Gavin Schmidt?

    You’ve already pulled the other one, change hands when you get to 100.

  26. Mark Hill Says:

    Graeme,

    The basics of sampling theory can be found in any introductory statistics book or on the internet.

    Why would any scientist of any persuasion seriously try to use a sample where n=5 ?

  27. JohnZ Says:

    Mark: They woudn’t.

    Archy is not a scientist, that paper obviously wasn’t peer-reviewed, and Graeme was taken in big-time.

  28. Mark Hill Says:

    Graeme,

    If you need help on basic science, maths, statistics, economics and finance, please visit the Khan academy.

    http://khanacademy.org/

  29. Graeme Bird Says:

    You are such an idiot Mark Hill. You will always be a moron.

  30. Graeme Bird Says:

    Everything I’ve said here is perfectly in keeping with sampling theory you dickhead. Fuck you are a dumb cunt Hill.

  31. Mark Hill Says:

    “Everything I’ve said here is perfectly in keeping with sampling theory you dickhead.”

    Oh really Graeme?

    n=5 is valid?

    How many regressors were there?

    Did you do a power test to prove this? Was it non-parametric for the highest level of robustness?

    No. I didn’t think so!

    This is like some disturbed wrtech dressing up in white pyjamas and thinking he’s the next Bruce Lee or Dolph Lundgren.

  32. Graeme Bird Says:

    Yes n=5 is valid. Because its cost-effective and you can always go back for more to enhance the statistical significance you dope.

    But if you use tainted data like everyone else did then you create a totally misleading impression.

  33. Mark Hill Says:

    “Because its cost-effective and you can always go back for more to enhance the statistical significance you dope.”

    Not actually doing to thest and firing the tax eater is also more “cost effective”.

    You muppet. You are completely out of your league. let the educated, civilised people continue the deabte.

  34. Graeme Bird Says:

    No you are a fuckwit mate. At no time have I gone against sampling theory. You will not find anywhere on this thread where I have contradicted sampling theory.

    You are just a moron mate.

  35. Jason Soon Says:

    n=1 is cost effective too turkey

    why don’t we just do future statistical studies with a sample of n=1?

    Study of insanity and utility of NZ immigrants

    sample n=1 Graeme Montgomery Bird

    We conclude from this study that the average NZ immigrant is insane and of low utility to Australian society.

  36. Mark Hill Says:

    “At no time have I gone against sampling theory.”

    You are a liar, mentally handicapped or both.

  37. jc Says:

    Sorry, Birdie, but you have lost this argument too.

    Birdie, it’s over, you have now officially lost every debate on this board. (are there others?).

    The only thing you can do is to apologize and go home. There’s nothing left for you to do.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: