NO. Please don’t take this as an insult, but your statements betray complete ignorance of the subject of statistical hypothesis testing.
Please. Stop bandying about phrases like “statistically significant.” You have no idea what you are talking about.
And I’m pretty confident that there exists no peer-reviewed paper that talks about statistical significance without (explicitly or implicitly) mentioning a null hypothesis.
Why is that? Generally speaking, a result is statistically significant if it is unlikely to have occurred if the null hypothesis is correct. That’s what statistical significance means.
brazil84: I suggest that you carefully re-read what I am saying. My point is that such statistical hypothesis testing in this case is very limited. All it would tell us is whether Hansen’s model is absolutely correct or not. As I already told you, I am ready to go out on a limb and say that it is almost certainly not correct. Actually, this isn’t that much of a limb because, given the current uncertainty in the climate sensitivity and the fact that Hansen’s 1988 model is toward the upper part of the probability distribution anyway, it is highly unlikely that there will not eventually be a statistically-significant deviation of the observed data from his model. (And, then there are the additional issues of assumed forcings and so on and so forth.)
That’s why such a binary either-or question is of limited utility. A better question is how well Hansen’s model does…i.e., how much does the data deviate one way or the other from it, not whether or not it is “perfect” in a statistical sense. (My estimate, without actually trying to do any sort of rigorous statistical analysis, is that at the moment it still is “perfect” in a statistical sense but that is because there has not yet been a large enough period of time to pin things down with all that much precision.)
LOL! I hadn’t done a search yet, but I suspected the topic had been discussed here. To someone like myself who’s followed the topic for years now, I keep seeing predictions (mainly on thunderbolts.info) of future events borne out and the astrophysicists scratching their heads saying they have no idea why their experiment turned out like it did. It’s hard for someone who thinks of science as mainly have value for turning out predictive theories/models to not give credence to a group making accurate predictions. I’ll go hunt down the previous thread on the subject and see what was said. I’m not really a proponent, but I’m obviously not a nay-sayer either.
And I would suggest that you take an introductory course in statistical hypothesis testing.
Again, your comments betray complete ignorance. A statistical hypothesis test NEVER reveals whether something is absolutely correct. There is ALWAYS some alpha error and beta error. (Actually it’s possible to completely eliminate one or the other by doing a meaningless test.)
Please please stop pretending that you know what you are talking about.
Following that link through told me that crank.net gave the theory a “Crankier” rating but not why, which is what’s frustrating when you’re trying to see if a theory is junk or not. The link on crank.net sent me to holoscience which sent me to thunderbolts.info, which I’ve been perusing for years and supports the theory as good science. So now new info for me there.
The link from TheFonz looks to be what I was looking for, thanks. I’ll peruse it some and see what that does for me.
-Eben
p.s. In reference to what I was talking about regarding the difference between proponents of EU and the critics is that EU supports say things like this while the critics seem content to say “Crankier” without explanation.
So I’ve read through the link some, and I hope like mad that I don’t read like TheFonz! I stopped about 2/3rds of the way through the first page 'cause I could see TheFonz wasn’t listening, just posting more semi-randomness. I’ve also seen much of the same arguments that he had, but I’m still thinking about them, or at least I think I’m thinking…
So there’s this little battle going on in my brain about what/who to beleive and I know that just because someone’s paranoid, it doesn’t mean there’s no one out to get them. In other words Al Gore and TheFonz (proxy of Thornhill maybe) are duking it out and they’re both rediculous, but what that say may have some truth at the core of it.
There was a link in that thread to an excellent refutation of much of the Electric Universe theory, so that was a good read.
I think this topic is pretty much done at this point. I’m going to mull over the EU arguments for a while and probably ask for a debate about some specific points later on.
Whatever. I admit that my word choice here hasn’t been quite optimal, since statistical hypothesis testing isn’t as big or formal thing in the physical sciences (or at least my part of them) as it is in fields where you don’t have much beyond statistics to go on. This has given you a good excuse to do what you do best…which is to pick on some tiny little point and ignore the bigger picture.
Anyway, I’ll note that this recent post at RealClimate is making, in another way, a similar sort of point to what I am making. It’s worth a read.
My point about null hypotheses was valid. You dismissed it from complete ignorance, and then shucked and jived for 5 posts or so rather than simply concede the point.
Anyway, small points are sometimes (but not always) important. When you start examining the details of CAGW theory, it becomes clear that the whole hypothesis is very shaky at best.
If it turns out that Hansen’s prediction was wrong, it’s a blow to CAGW theory, since he made his predictions without the benefit of the last 20 years of temperature records.