[quote]Bill Roberts wrote:
So you’re saying, if they are correct that the consequences are dire, then the consequences are dire?[/quote]
I’m saying that if they are correct about their ecological predictions then the consequences of not enacting economic regulation–those ecological predictions–are dire. So sure.
[quote]Bill Roberts wrote:
[quote]stokedporcupine8 wrote:
[quote]PRCalDude wrote:
[quote]stokedporcupine8 wrote:
While Bill’s criticisms are not without substance, this still sounds a bit simplistic to me. While it seems odd that you can’t give accurate short term predictions but can give accurate long term predictions, it’s not hard to imagine why. The two sorts of predictions are dependent on different variables. Short term predictions require knowing all sorts of precise factors, while long term trends (according to geological records) seem to be fairly correlated to only a few easy to measure variables. To blindly assume that being able to predict one allows you to predict the other is to miss how the predictions are made.
[/quote]
Do you have any evidence of this - that “long term predictions are correlated with only a few easy-to-measure variables?” Why would short-term measurements be any different? You seem to accept this based on what someone else has told you.
The fact that long term geological trends are correlated with certain variables does not mean that those variables caused the trends, rather those variables provide the most accurate data ABOUT the trends.
If what you are saying were true, scientists would have no reason to fudge the data they got from their models, right?
Just out of curiosity, how much modeling have you actually done?
[/quote]
I’m not a climatologist, I’m simply mentioning important distinctions that Bill has not mentioned in his criticism. [/quote]
Well, in this particular thread I’ve preferred to make only fairly short statements.
More broadly speaking, I have addressed this at other times and you’re right, it should be addressed again.
To evaluate a model legitimately, one has to see if it is capable of predicting data that was not in any way used to create that model.
It is no feat whatsoever to create a model that matches, reasonably closely, known data. For example, you could provide a string of random numbers and a function can be created that fits that series exactly.
But any claim that that function is capable of predicting future numbers would be completely unwarranted and false.
Or you could have some very complex function which generates a series of numbers from some inputs, and you could provide me with some subset of that output and tell me what some or even all of those inputs are. It would be no great feat to produce a model which on those given inputs and known outputs produces an almost-matching series. But this would not mean that the model actually replicates your very complex function – it’s quite unlikely that it will: it may in fact completely lack or overweight perhaps some feedback functions or dependencies on given variables – or that it will continue to produce matching outputs.
It may well only be able to match the outputs of your function where those outputs were used in creating the model. For the model to be taken seriously, it would have to be able to predict data not used to create the model. Without that, it’s an exercise and nothing more, and most certainly it proves nothing.
If modeling as done by the AGW crowd were done in an intellectually honest manner, their demonstration of validity would include that the model in fact “predicts” known data that was not used in the creation of the model.
That hasn’t been done. And when it has been tried it fails: e.g., when you leave out the 1950s, the models cannot “predict” the 1950s.
I recall the predictions made around the year 2000 for the 2000-2009 decade. The temperature projection kept rising past the 1998 level.
The model did not predict correctly.
What the AGW computer modeling is is nothing but fitting the same data that was used to create the model and nothing more.
This is bullshit and would not, so far as I know, be accepted as proving anything in any other field.
[/quote]
Yes, I don’t know how I got sucked into this. I don’t know enough about these things (what sorts of predication have been made, how they were made, etc) to get into an argument about them.
I also, as I’ve mentioned in other threads, try at length to differentiate between the political activist nut jobs like Gore and working climatologists. While I’m sure it is quite easy to pick apart the radical claims of those political activist nut jobs, the fact that so many working climatologists in some form or another think that there are potentially global consequences to our economic activities is not so easy to dismiss.
(Oops: either you added “ecological” in an edit, or I failed to notice it before – no need now for my post clarifying why the statement seemed a tautology.)
Peter Schiff is a stopped clock and has not made any good predictions. I’ve posted on that topic before.
How do I know capitalism is good? Well, what has worked better? How did fascism and communism work out last century? As far as macro-economic theories are concerned, there are none with any predictive value as near as I can tell. Even the Austrians are all over the map with their predictions about our money supply expansion and what it means for inflation. Microeconomics is basically codified common sense, so there’s not reason really to disbelieve it.
Did you know that the guy who created the model for housing defaults used to securitize mortgages was a PhD-level statistician? His model was based on things that he thought were correlated that were, in fact, uncorrelated:
I guess my point is that very smart people make very dumb assumptions all the time. When they gain a lot of influence and are in charge of a lot of money, they can make everyone else’s life hell also. So it was with the Minority Mortgage Meltdown, so it will be with the nascent cap and trade market.
I say that there’s no science whatsoever going into AGW modeling. The whole thing is either a) a religion and/or b) an effort to make the right people a lot of money, just like we saw through much of this decade.