Now you and the other AGW theists can go out and sacrifice a virgin to appease the volcano gods.
Posted on 05/03/2008 8:58:57 PM PDT by cogitator
An analysis of recent temperature data by two scientists at the Australia Bureau of Meteorology.
Waiting for Global Cooling (PDF)
Doomage.
wha’ i can’t wait to see
is how the u.s. democrap media
spins the brains of tv viewers
from global warming cycle
to global cooling cycle!
si.
He reports a linear upward trend in temperature on page 4 but does not provide the statistics that tell you how likely it is that the trend is different than zero. I ran the actual regression numbers for two of the series, the one with the strongest trend (NASA GISS) and the one with the weakest trend (HadCRUT3). The results appear below after the text.
The bottom line is that neither of the "trends" comes even close to the normal p-value required to be considered statistically different from NO TREND. Generally p<0.05 is regarded as the threshold. His p-values, had he reported them, would have been about 0.17 (HadCRUT3) and 0.67 (NASA GISS).
This omission is really embarrassing--especially when reporting the p-values makes his "trends" meaningless. From glancing through the article, he almost certainly knows enough statistics to know that his "trends" are statistically meaningless and he almost certainly knows better than to compute a regression "trend" and not report the p-value for the regression. That he did not was, imho, almost surely intentional.
In formal terms, none of the series reported on page four of the article give any reason to reject the null hypothesis at the 95% (the standard), the 90% or even the 85% confidence level. The null hypothesis would be that there is NO trend in temperature between 1998 and 2007.
If this were a normal scientific paper, the reviewers would have required him to report that, as a result of his data, he could not reject the null hypothesis that there is NO TREND in temperature.
*******************
For hadCrut3 Series
TEMP = a*Year + b
Coefficients:
Value Std. Error t value Pr(>|t|)
(Intercept) -7.7865 18.6444 -0.4176 0.6872
Year 0.0041 0.0093 0.4396 0.6718
Residual standard error: 0.08852 on 8 degrees of freedom
Multiple R-Squared: 0.02359
F-statistic: 0.1933 on 1 and 8 degrees of freedom, the p-value is 0.6718
********************
For NASA GISS Series
TEMP = a*Year + b Value Std. Error t value Pr(>|t|)
(Intercept) -39.6009 27.0090 -1.4662 0.1808
V2 0.0204 0.0135 1.5092 0.1697
Residual standard error: 0.1282 on 8 degrees of freedom
Multiple R-Squared: 0.2216
F-statistic: 2.278 on 1 and 8 degrees of freedom, the p-value is 0.1697
Just one more note. What is also interesting in this paper is that noone is discussing the fact that the 1999 and on numbers (which this AGW’er accepts as correct) are way below the IPCC forecasts.
Assessing just how badly the IPCC did is difficult because of the sloppiness of the IPCC forecasts—you really can’t tell what the meaning of their “confidence intervals” are. But if they are typical confidence intervals (95%) bands, the probability that the IPCC forecasts are too high is very large (I ran those numbers about a month ago and was suprised that noone is talking about just how far off the IPCC was.)
Even if you toss the IPCC confidence intervals and use the IPCC temperature forecasts only, it is still highly probable that the IPCC forecasts are too high.
I’ll try to find these numbers and post them here.
Sloppiness or willful distortion?
Syria reached -22F this winter and Antarctica is 2C below the mean. Do you want to hear about China's recent coldest winter in 100 years?
Oh sure, go ahead and believe facts. I’ll stick with Algore (Nobel Prize winner don’t you know).
If I’ve said it once, I’ve said it a hundred times: Lower temperatures are a classic sign of global warming.
Do the analysis starting in 1999. Because:
"In other words, the reason that 1998 was so exceptionally warm is that a very strong El Niño interacted with the global warming trend to give an exceptional year."
How does the observation that March 2008 was the second-warmest all time (warmest ever over land) jive with that?
And are you sure about Hansen's data for the 1920s? I thought it was the 1930s.
Complete baloney! Your post the other day stated that in the USA March was the 63rd warmest in the past century. That's cold. As for the temperature data set for the rest of the world NOT matching the trend in the USA for March I call it suspect.
Now that we have the Argo network in the oceans and the Aqua satellite the truth is being shown and that is the Earth is cooling.
I’m sure glad that we are not intending on using crops for fuel... Oh, wait a minute......
If he then reports the statistically significant slice relationship without performing what is called the Bonferonni adjustment, he is is in a state of statistical sin.
In the data at hand, what you have just done is data-dredging. You have picked the period by eye that is most likely to show a relationship and want to know the numbers for that time slice. Because you just did a pretty good job of picking one of the most favorable for your hypothesis, that is the same as if you ran the numbers on all possible time periods for all possible series.
Off the top of my head, there are 5x5x5 possible series of five or more points to report in the author's data. That's about 125 different "slices" you could test (I limited it to five to help your cause, you are unlikely to get a statistically significant regression out of fewer than five). You just picked one of the most favorable of those 125 slices. But if I ran all 125 slices at the 95% confidence level, about 6 of them would show a statistically significant relationship sheerly by random chance.
The question at hand is not, "can I slice the data so I can report a statistically significant relationship that is consistent with my hypothesis." It is, "does my original data support my original hypothesis at my originally chosen confidence level?" That's why you define your test, your significance level, and the data in advance. It avoids the sin of data dredging.
The scope of my response was limited to the author's choice of data and his hypothesis (note, I couldn't use his significance level because he didn't report it). That let me avoid dredging the data and other related sins such as adjusting your significance level downward once you see the data.
So, with that caveat, I ran things somewhat sloppily but I'm pretty sure the results are: You can find a statistically significant upward trend only in one of the three series: NASA GISS and only on a few of the 125 slices. I'm pretty sure out of the 125 total slices, you have 4 slices (one of them the one you requested) that are statistically significant at the 95% level. Compare that to the expectation that 6 of those 125 slices will show a statistically significant relationship by random chance.
So even done your way, the overall data set is consistent ONLY with the hypothesis of NO TREND.
Does that mean your la nina ended or did it take a vacation for a month?
Run the numbers anyway. With an original p of 0.67 for the one data set, there’s probably a good chance that p will still exceed 0.05.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.