--- John Foster <[log in to unmask]> wrote:
> >> I have performed a correlation of the temperatures measured between
> 1932
> >> and
> >> 2000 to see if there is a linear trend of 'warming' at Churchill,
> >> Manitoba.
> >
> >ROFLMFAO!
> >
> >Look after John excoriates me for using a linear model with multiple
> >variables, John uses the most brain dead simple model possible, the
> linear
> >trend model. A model that is so patently unreal that it isn't taken
> very
> >seriuously except for short term forcasting in limited instances and to
> >provide comparisons to more sophisticated models. LMAO. Hooo, hooo it
> is
> >too funny.
>
> How else are we to prove that your assertion was wrong again...
>
> No you are too funny. You commented in an earlier post that the
> temperature
> in Hudson Bay was getting cooler. I took all the temperatures recorded
Yes it is obvious if you look at the data I provided a link to.
> for
> the months of January and February from 1932 until 2000 to see if it was
> getting warmer. It has been getting warmer. I did not use linear
> regression,
> but rather made a test of correlation concerning the observed
> temperatures
> and month of year. The scatter plot shows that it is warming [Pearson
It is essentially the same thing John. You were looking to see if there
is a linear trend in the data over time. Sheesh. What you are saying is
that over time temperatures have warmed. You can't say why or if it will
continue. Your analysis is even more simplistic than mine.
> Products method]. There was no attempt to make any inferences as to
> cause
> using the test of correlation. However I did provide a 'casual'
Yes I know, you noted a correlation then come up with a song and dance.
No attempt to see if your explanation has any measurable support.
> explanation
> as to why the temperatures have risen since 1932. The correlation proves
> 'Arctic warming' is occurring...
Well, no. It proves that there has been warming. Further a casual
inspection of the data will show that your correlation is the result of
the data at both ends of the series. Doing a correlation an dropping the
first and last 10 years will change the results dramatically. In fact,
let me do the analysis right now....uh oh, your correlation just dropped
to
0.13062
By the way, I did a linear regression with temps as the dependent variable
and time as the explanatory variable for January.
The mulitple R: 0.23716
Pearson's Product Method:0.23716
Well I be damned. Same damn number. Well what do you think this means.
How about that John is essentially doing a linear regression. The Pearson
Product Moment Correlation measures the linear relationship between two
data sets.
> Now I will ask you to demonstrate or prove that temperature in Churchill
> is
> getting cooler....but you cannot.
I didn't say that temperatures were cooling in Churchill, but that
temperatures in the Hudson Bay area were predominantly flat at each
station, and cooling out on the Bay itself. That is what the data I
linked to says.
Steve
=====
"In a nutshell, he [Steve] is 100% unadulterated evil. I do not believe in a 'Satan', but this man is as close to 'the real McCoy' as they come."
--Jamey Lee West
__________________________________________________
Do You Yahoo!?
Yahoo! Shopping - Thousands of Stores. Millions of Products.
http://shopping.yahoo.com/
|