Recommended Posts
jcd11235 0
brenthutchWhen a models predictions do not comport with observation (aka reality) the model is WRONG.
That is incorrect. Predictive models are not expected to correctly predict future observations 100 percent of the time. Increasing confidence levels for prediction intervals results in wider intervals, decreasing the overall usefulness of the models.
A weatherman could say, with 100 percent confidence, that the high temperature tomorrow will be between -460º and 9941º F. He would be guaranteed to be correct, but his prediction would be completely useless. He might only be 90 percent confident that the high temperature will be between 75º and 85º F. (Note: I'm not a meteorologist, so I don't know the actual width of a 90 percent confidence prediction interval provided by current weather models.) While the smaller prediction interval is expected to cover the actual high temperature only nine out of ten times, it is a far more useful prediction.
If we monitored the weatherman's predictions (at the 90 percent confidence level) over an extended period of time and found that the intervals only covered the observed values 60 percent of the time, we would be justified questioning the accuracy of his model. A model that provides the expected proportion of accurate predictions is not wrong.
Math tutoring available. Only $6! per hour! First lesson: Factorials!
kallend 1,672
winsor******Don't take my word for it. Look at the IPCC's models and then look at NOAA's observations. When a models predictions do not comport with observation (aka reality) the model is WRONG. It does not matter how big of an asshole the critic is or how saintly the advocate is, the theory is WRONG. Grow some balls and some brains and get over it.
So you're still having trouble telling the difference between climate and weather.
f=ma is wrong too. But it's still better than no model at all and allows us to design airplanes, helicopters, ships, trucks and cars.
E=mC^2 +/-3dB
Yes, but very difficult to use in airplane design.
...
The only sure way to survive a canopy collision is not to have one.
The only sure way to survive a canopy collision is not to have one.
D22369 0
Also, my balls and brains are just fine, but it warms my heart to hear you're thinking of them. Also, I'm married. Sorry.
***
ahhh, yer married, did Errin get a coin purse to hold em in? and you think too much
Roy
Ps: hope to see you guys soon, been a while
They say I suffer from insanity.... But I actually enjoy it.
So you're still having trouble telling the difference between climate and weather.
f=ma is wrong too. But it's still better than no model at all and allows us to design airplanes, helicopters, ships, trucks and cars.
E=mC^2 +/-3dB
Share this post
Link to post
Share on other sites