Posted by Lowell on December 07, 2001 at 15:28:38:
What is the difference between a miss and a false alarm, then? My definitions come from a Master's thesis done at Georgia Tech by Dr. Fred Creamer. I don't much care what words we use, but it is best if we can agree on definitions before we try evaluating forecasts en masse. Would anyone care to write definitions and give words to the following occasions: 1) An earthquake occurs but was not forecast. 2) A forecast was made but no corresponding earthquake occurred 3) A forecast was made and an earthquake completely matching the forecast occurred, but the odds of random success were greater than 1 in 3 (or whatever odds you chose for a successful prediction which would most likely have occurred no matter whether the predictive method had any validity or not. 4) A forecast was made and an earthquake occurred which partially matched the forecast parameters and the odds of occurrence were less than 1 in XX i.e. a forecast which beat the odds of success considerably but was not an exact match with the event. 5) A forecast completely describes a subsequent earthquake in all particulars and the odds of random success are less than 1 in XX Any other potential possibilities you can think of. We need words to these whatever those words are, so we can describe these occasions when words are necessary that we can all understand when we use them.
Follow Ups:
● Re: False Positives - bobshannon.org 17:17:09 - 12/7/2001 (11726) (2)
● Re: False Positives - bobshannon.org 23:08:03 - 12/7/2001 (11739) (0)
● Re: False Positives - Roger Hunter 17:44:06 - 12/7/2001 (11732) (0)
● definitions - Roger Hunter 16:58:07 - 12/7/2001 (11723) (0)
● Re: False Positives - Canie 16:18:06 - 12/7/2001 (11721) (1)
● Re: False Positives - bobshannon.org 17:20:05 - 12/7/2001 (11727) (0)
|