[Roger Hunter]

(03-04-2017, 07:45 PM)Duffy Wrote: Duffy;

Quote:With respect, it would ordinarily make no difference where you sourced your selection from, personally a few numbers off a calendar would have sufficed.  But if you have indeed chosen post seismic locations for your dates, it not only contradicts the idea of the test, it also enforces my hypothesis of quake to quake propagation.

No, I picked those date out of the air. If they coincided with yours it was only coincidence.

Quote: Firstly, I have on several occasions postulated the idea of energy transfer between quake locations. If you analyse these very results, you will see that during multiple events, the terminator threshold's were in the proximity of recently active earthquakes, whilst a present quake was occurring ... and you got a hit!   We spoke of this at some length 18 months ago, I'll try and source the thread so you can refer to it again. I don't generally speak of these things anymore because I didn't do me any favours back then, nor I believe will it do me any favours in the future ... unless of course I am able to predict earthquakes.

Exactly; no credit until success is proven.

Quote: Secondly, I am testing a signal source that I am trying to determine if there is correlation with earthquakes.  If my first point is anything to go by, then it stands to reason your selection will undoubtedly score high. If my source does have correlation with recent events, and you chose resent events, then your crosses are going to connect with mine.  This is the point I was trying to make by showing association, the margin used was 0.75 degrees per threshold, giving a total band of 3 degrees on a global scale.   It is inevitable that some association will occur (you and I hit sunrise at the same time etc), but this happened with 50% of your results.  Without further tests, we can only speculate that this is a possibility every time.  However, as mentioned I have run multiple simulations ... it did not happen to this extent on these occasions, but there is only my word for this.

I think it's just the short duration of the test. I gave no attention to your list but there are only so many days to choose from.
 

Quote:I have no problem with you giving your opinion or conclusion with the results at hand.  It gives indication of one's standing with respect to the project, but you always seem to give a final, definitive, no point going on answer which is very hard to get use too.

Perhaps I'm just being too brief in my responses. No offence is intended.

Quote: You cannot test this like I can, you haven't got your program running, and we seem to have a new definition of "random" on Earthwaves ... which I can live with.  It would be nice to try some numbers off a calendar next time, maybe then you will see the difference.

You lost me there. Dates are dates, calendar or not. And once I get the bloody program running correctly, I'll be able to test a huge number of cases and give a definitive answer.

Quote:I think I should point out, I am assuming that I maintain the right to scrutinize yours and Brian's results, just as you have the right to scrutinize mine.  I mention this because I don't want to be seen as being offensive when in fact I am defending my hypothesis ... I don't want to take on the boss and his foreman, if it means I'm going to lose my job at the end of it!  

Not to worry, there are no authorities here. (Except Brian; he can throw us both off if he chooses!)

Roger