I can find quite a bit of fault with a 3 month statistical data set when dealing with such geological data as earthquakes.

3 Months is but a moment of time - re-do your data for the past 4-5 years and you may get a better data set. Statistically speaking.

You also do not state your methods of what you consider an 'accurate' pediction - was there any room for error ? a few hours +/- or a few miles +/- ?? - We do not have a stated method on this site of how to accurately state a prediction as in Jim's board - what did you consider a prediction ? Very few of our 'normal' posters have been posting here lately (the board hasn't even been around for 3 months at this site) - there's been too much 'noise' in the wings around here and some of our best predictors (such as Dennis, Dyan, stargzr) aren't posting.

You completely mis-understood Diane's quake 'warnings' - she has windows of probability of a quake and her method is more of an anti-quake sort of think - that a quake will only happen in her windows - not guarentee it will happen in that time frame - Her certainty is more along the lines that if there is not an open window, a quake CAN'T happen...

Anyone can prove anything statistically using scewed data sets.

Canie