comp.lang.idl-pvwave archive: archive » Re: Statistic codes: Significance level

Home » Public Forums » archive » Re: Statistic codes: Significance level

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: Statistic codes: Significance level [message #38274]

Tue, 02 March 2004 12:44

wmconnolley
Messages: 106
Registered: November 2000

Senior Member

In article <4044FD8C.9020906@ya.com.spam> you wrote:
> First of all, thank you Bob, Craig, and William for the direction.

> I am now looking at each suggestion to see which one(s) fit(s) the
> statistics I am asked to perform (I am still kind of a beginner
> with respect to massive data statistics). I will try to sumarize
> my findings once I get some results.

>> ; for i=1,ntest do a=reg_explain(pp_regress(x,randomn(seed,n) $
>> ,/arr),prob=.1,/qu)+a

> With respect to the suggestion made by William Connolley, I am
> afraid I don't know what PP_REGRESS is. I beleive it should be
> another function which returns some Percentil-Percentil result
> or something similar.

Hi. Sorry about that... pp_regress is of course one of mine.
Basically it just wrap up REGRESS because I could never remember
how to use it, and works for a particular type of field too
(pp-fields). Find it at:

http://www.antarctica.ac.uk/met/wmc/idl/pp_regress.pro

You'll want to use it with arrays (not structures) and use /array
for your purposes. The line above is just a Monte-Carlo thingy
to test that the statistics coming out are about correct (a useful
test for those of us who can't do Big Maths since our brains
decayed).

BTW, if anyone is interested in break statistics in time series
(Lund/Wang) see:

http://www.antarctica.ac.uk/met/wmc/idl/lund/

-W.

--
William M Connolley | wmc@bas.ac.uk | http://www.antarctica.ac.uk/met/wmc/
Climate Modeller, British Antarctic Survey | Disclaimer: I speak for myself
I'm a .signature virus! copy me into your .signature file & help me spread!

Report message to a moderator

[Message index]

		Re: Statistic codes: Significance level By: wmconnolley on Tue, 02 March 2004 12:44
		Re: Statistic codes: Significance level By: Andry William (Please on Tue, 02 March 2004 13:33
		Re: Statistic codes: Significance level By: Craig Markwardt on Tue, 02 March 2004 01:32
		Re: Statistic codes: Significance level By: wmconnolley on Tue, 02 March 2004 01:45
		Re: Statistic codes: Significance level By: R.G. Stockwell on Mon, 01 March 2004 16:35

Previous Topic:	2D-fit
Next Topic:	Re: 2D-fit

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Fri Nov 28 09:29:29 PST 2025

Total time taken to generate the page: 0.00809 seconds