Belmont Club

By Richard Fernandez

Bio

Get Updates From Richard Fernandez
A Comment About

Ho ho ho

July 27, 2008 - 6:49 pm - by Richard Fernandez
fred
2008-07-27 21:13:14

‘The question is whether we can predict the change from our models. Whether our models add any value to the prediction.”

If we get beyond five days, weather models experience considerable degradation in their accuracy of prediction. I am not aware of any reliable and decently accurate climate models, but if there are I would be the first to listen. Yet, I just cannot fathom how even a computer and algorithms could possibly take into account all the variables and especially how everything in the timing of events can change when an unexpected variable intrudes. Call me a skeptic. Anyway, there are too many variables, and no one factors in the sun’s behavior and when it will come into play. Ditto for the seismic, deep ocean events which release gargantuan amounts of thermal energy and this eventually effects surface temperatures of the ocean even if only one degree. Can mean the difference in ocean currents and wind patterns.

I’m not against the effort to try. I just am in awe of the task. If we in the finance and investing world find that our models can fail and do in fact fail – with actually fewer (even if human)variables to account for in the algorithms – how much more daunting is the task when modeling nature?