[Roger Hunter]

(11-27-2016, 08:43 AM)Duffy Wrote:

(11-26-2016, 05:37 PM)Roger Hunter Wrote:

(11-26-2016, 04:00 PM)Duffy Wrote: Roger

Your conclusion thus far is troubling in respect that the Uni may also come to the same conclusion.  Period of prediction window and the rate of posted predictions, will always be in my favour until the data used to procure said predictions is analysed accordingly.  This chicken and egg scenario is getting very tiresome, all I can do here is give you numbers because you don't have access to the data, and the professor I had in my very home would not look at the data until he had seen the numbers !    As a non scientist, I am finding it difficult to understand why I could give advanced notice (twice) of something imminent occurring in New Zealand.   Then it being changed into a number, which would then suggest I would have probably got it anyway !  In truth, I was 4 degrees off, but then again, I only have a laptop screen and a felt pen. I eventually got my prediction on 177' E, only to be reminded by you that " predicting moderately sized quakes in seismic regions will not count for much", in England we call this "rubbing salt in the wound".

On the three consecutive days of 16th, 17th, 18th November, I recorded very strong signals which I calculated as being Mid-Atlantic Ridge at 38' 17' W ( opposite being 141' 43' E .... 6.9 off coast of Namie, Japan ! ). I posted Mid-Atlantic prediction on the 19th.  As my hypothesis suggests an opposite reaction, I searched my data for a correlating signal and found a match on 8th Nov. It was a 07:30 ut sunset signal for 138' 44' E by 43' 56' N  Off West coast of Hokkiado, Japan. I confirmed it as a seismic anomaly with ACE data that I had saved from this period.  I could not cross reference this with a sunrise signal because someone keeps stopping the satellite feed at the crucial time coverage is needed, but I predicted anyway.

The 6.9 quake off Namie ,Japan happened on 141' 24' E by 37' 22' N ... Sunset at this exact bearing on 8th Nov was 07:30 ut.  Both co-ordinates shared the same Sunset time on the 8th, easy enough to determine after the fact, but what are the odds of finding this before the event occurred ? ... I spend hours and hours analysing data for predictions, and you people just turn it into a number, and it becomes meaningless !  Until someone checks my data, the odds will always be in my favour for the wrong reason.  To conclude this, a 4.7 occurred in the Mid-Atlantic on the 24th on 41' W ( opposite being 139' E ), but being a 4.7, it has little relevance here.

Ok Roger, your conclusion shows the odds of achieving any results of significance in the future is at best "low".  I cannot continue predicting in this manner with little or no chance of success, so I propose my hypothesis be tested under different circumstances, to those under normal prediction protocol. 

If two corridor's of longitude were selected, each at 10 degrees of lateral width ( including opposite Corridor's). Then when a possible 5+ is detected in said corridor's, I place a prediction in longitude with a 2 degree margin and a one month window. Would this be of significance if I got a positive result ? ... or would it be seen as a parlour trick ... you get to choose the corridor's ! 

Duffy

Duffy;

I understand your frustration but I can't do anything to resolve it. The main problem is your 30 day window for small (5+) quakes. It's impossible to fail with that, they happen nearly every day.

You might modify that to say you have a signal which indicates that a mag 7+ quake will happen somewhere and it will be either dawn or sunset when and where it happens. That's testable since 7+ quakes are less frequent.

Your alternate proposal won't work. I can select corridors that never have quakes.

The ideal prediction gives location, date and magnitude, all correct within reasonable limits, So far, no one can do this consistently. Lucky guesses are not enough.

What I would like to have is a list showing only the date/time of your signals. I can compare that to the quake list to see if there is anything consistently happening to indicate a correlation.

Roger

If I stay on this course ... I'll end up like Ritcher

Your ok too, Roger

Duffy

Thanks Duffy, I try.

Another quick test is to see if quakes happen at sunrise/sunset, moonrise or moonset. I can do the sun easily, it's just local time but moon is harder.

I'll let you know. Working on it (sun) now.

Roger