[Roger Hunter]

(03-19-2016, 04:01 PM)Duffy Wrote: Hi Roger,

I appreciate the advice, but what you are suggesting is not achievable yet with the equipment I have ... a 7 on my screens would be no more discernible than a barrage  of 5's !!, I am working on this, but aspect changes in the data week by week are not making it easy.

As you didn't answer my question about high probability, I can only assume you though I was being satirical again when in fact I was trying to understand the logic in what you said, I was not aware of a different level of hit which is dependant on event region .. if this is the case and you have published something here that I have over looked .. I apologise for the question, The hits are of little consequence but it is nice to have something to aim for, my English manner I guess !.

Duffy,

Duffy;

Ok, so all you know from your screens is that something is happening (or likely to happen) somewhere soon (maybe).
That makes it tough to proceed.

The best you could do is state that some sort of seismicity can be expected somewhere in the near future, which is always true.

The probability question works this way in my evaluation program; when a specific location is predicted I check the history for that area to get the probability of a hit. This is used to assign a value to the results. A miss counts against you, a hit counts for you. The highest score would be to predict (correctly) a quake where none has ever happened. The lowest score would be for a location which has them all the time. Magnitude is part of the analysis, large one being less common.

Date predictions consider only date and mag. and get lower scores accordingly.

Roger