(03-28-2014, 12:13 AM)Roger Hunter Wrote: If you have an idea that quakes are triggered by the full moon or whatever, let me know and I'll test it for you.

It has to be something involving data that's available online, like planetary positions, tides, things like that and you have to be able to put it into rules.
IF this happens THEN a quake of given size will happen THERE. That sort of thing.

So far I've tested Berkland, Petra, Shan, Shou and Amit.

No charge; it's my hobby.

Roger

(04-22-2014, 12:03 AM)Roger Hunter Wrote:

(03-28-2014, 12:13 AM)Roger Hunter Wrote: If you have an idea that quakes are triggered by the full moon or whatever, let me know and I'll test it for you.

It has to be something involving data that's available online, like planetary positions, tides, things like that and you have to be able to put it into rules.
IF this happens THEN a quake of given size will happen THERE. That sort of thing.

So far I've tested Berkland, Petra, Shan, Shou and Amit.

No charge; it's my hobby.

Roger

Amit has a theory that large aftershocks will happen at 12 hour intervals
or 7 1/2 days after the main shock.

I looked at all mag 6+ quakes in the neic catalog (1973-2013)
The program read thru the file looking for a mag 7.5+ quake.
When it finds one it starts computing the hours until the next quake.
This continues until it finds a new main quake or until 8 days have passed.

Code:

HOURS AFTER THE MAIN QUAKE
| 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
|-----------------------------------------------------------------------
|86 30 22  7  5 10  6  6  7 12  9  6  7  6  8 12 13  1  6  2  5  6  6  8
| 8  6  3  3  5  5  2  5  5  6  2  4  3  5  2  3  4  3  5  3  2  4  6  1
| 1  4  1  2  5  3  5  1  3  5  2  5  2  1  1  1  5  1  2  0  2  6  2  1
| 1  1  2  6  2  2  1  2  5  3  7  1  3  3  3  1  5  0  1  4  1  4  2  4
| 4  2  1  6  2  1  2  3  3  3  5  7  4  2  0  5  1  1  2  1  5  5  4  6
| 6  2  4  1  5  3  3  3  3  0  4  4  7  2  3  1  7  2  2  5  5  4  3  1
| 1  5  1  3  6  5  2  5  4  0  2  5  1  1  1  2  2  5  0  2  1  2  3  6
| 6  3  4  3  0  1  0  3  4  1  4  2  2  2  4  2  4  0  2  3  3  4  1  2

Looking at the table you will see that there is no peak at 12 hours
and none at 7 1/2 days.

Roger

(04-23-2014, 05:19 PM)Amit Wrote:

(04-22-2014, 12:03 AM)Roger Hunter Wrote: [quote='Roger Hunter' pid='208' dateline='1395965631']
If you have an idea that quakes are triggered by the full moon or whatever, let me know and I'll test it for you.

It has to be something involving data that's available online, like planetary positions, tides, things like that and you have to be able to put it into rules.
IF this happens THEN a quake of given size will happen THERE. That sort of thing.

So far I've tested Berkland, Petra, Shan, Shou and Amit.

No charge; it's my hobby.

Roger

Amit has a theory that large aftershocks will happen at 12 hour intervals
or 7 1/2 days after the main shock.

I looked at all mag 6+ quakes in the neic catalog (1973-2013)
The program read thru the file looking for a mag 7.5+ quake.
When it finds one it starts computing the hours until the next quake.
This continues until it finds a new main quake or until 8 days have passed.

Code:

HOURS AFTER THE MAIN QUAKE
| 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
|-----------------------------------------------------------------------
|86 30 22  7  5 10  6  6  7 12  9  6  7  6  8 12 13  1  6  2  5  6  6  8
| 8  6  3  3  5  5  2  5  5  6  2  4  3  5  2  3  4  3  5  3  2  4  6  1
| 1  4  1  2  5  3  5  1  3  5  2  5  2  1  1  1  5  1  2  0  2  6  2  1
| 1  1  2  6  2  2  1  2  5  3  7  1  3  3  3  1  5  0  1  4  1  4  2  4
| 4  2  1  6  2  1  2  3  3  3  5  7  4  2  0  5  1  1  2  1  5  5  4  6
| 6  2  4  1  5  3  3  3  3  0  4  4  7  2  3  1  7  2  2  5  5  4  3  1
| 1  5  1  3  6  5  2  5  4  0  2  5  1  1  1  2  2  5  0  2  1  2  3  6
| 6  3  4  3  0  1  0  3  4  1  4  2  2  2  4  2  4  0  2  3  3  4  1  2

Looking at the table you will see that there is no peak at 12 hours
and none at 7 1/2 days.

Roger

Quote:Roger
I can not make out anything from this. Are thse figures pertaining to a single quake of 7.5?
How many quakes of more than 7.5 have to plotted ?
Amit

(04-23-2014, 06:21 PM)Roger Hunter Wrote:

(04-23-2014, 05:19 PM)Amit Wrote:

(04-22-2014, 12:03 AM)Roger Hunter Wrote:

(03-28-2014, 12:13 AM)Roger Hunter Wrote: If you have an idea that quakes are triggered by the full moon or whatever, let me know and I'll test it for you.

It has to be something involving data that's available online, like planetary positions, tides, things like that and you have to be able to put it into rules.
IF this happens THEN a quake of given size will happen THERE. That sort of thing.

So far I've tested Berkland, Petra, Shan, Shou and Amit.

No charge; it's my hobby.

Roger

Amit has a theory that large aftershocks will happen at 12 hour intervals
or 7 1/2 days after the main shock.

I looked at all mag 6+ quakes in the neic catalog (1973-2013)
The program read thru the file looking for a mag 7.5+ quake.
When it finds one it starts computing the hours until the next quake.
This continues until it finds a new main quake or until 8 days have passed.

Code:

HOURS AFTER THE MAIN QUAKE
| 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
|-----------------------------------------------------------------------
|86 30 22  7  5 10  6  6  7 12  9  6  7  6  8 12 13  1  6  2  5  6  6  8
| 8  6  3  3  5  5  2  5  5  6  2  4  3  5  2  3  4  3  5  3  2  4  6  1
| 1  4  1  2  5  3  5  1  3  5  2  5  2  1  1  1  5  1  2  0  2  6  2  1
| 1  1  2  6  2  2  1  2  5  3  7  1  3  3  3  1  5  0  1  4  1  4  2  4
| 4  2  1  6  2  1  2  3  3  3  5  7  4  2  0  5  1  1  2  1  5  5  4  6
| 6  2  4  1  5  3  3  3  3  0  4  4  7  2  3  1  7  2  2  5  5  4  3  1
| 1  5  1  3  6  5  2  5  4  0  2  5  1  1  1  2  2  5  0  2  1  2  3  6
| 6  3  4  3  0  1  0  3  4  1  4  2  2  2  4  2  4  0  2  3  3  4  1  2

Looking at the table you will see that there is no peak at 12 hours
and none at 7 1/2 days.

Roger

Quote:Roger
I can not make out anything from this. Are thse figures pertaining to a single quake of 7.5?
How many quakes of more than 7.5 have to plotted ?
Amit

No, the table is the cumulative results of all 7.5+ quakes from 1973-2013, looking for anything smaller up to 8 days later. There were several hundred of them I'll go back and count them if it is important.

Roger

New test; Amit has asked that people try to beat his success ratio.

I used as my data set the mag 6+ quakes from 1993-2013.
The odds on getting a hit in a 3 day window are 0.621.

I picked 600 random dates for that time, requiring each one to be at least 3 days away from any other selected date and checked to see if the corresponding window contained any quakes.

Results to follow.

Roger

---

(04-24-2014, 01:09 AM)Roger Hunter Wrote:

(04-23-2014, 06:21 PM)Roger Hunter Wrote:

(04-23-2014, 05:19 PM)Amit Wrote: [quote='Roger Hunter' pid='250' dateline='1398124989']
[quote='Roger Hunter' pid='208' dateline='1395965631']
If you have an idea that quakes are triggered by the full moon or whatever, let me know and I'll test it for you.

It has to be something involving data that's available online, like planetary positions, tides, things like that and you have to be able to put it into rules.
IF this happens THEN a quake of given size will happen THERE. That sort of thing.

So far I've tested Berkland, Petra, Shan, Shou and Amit.

No charge; it's my hobby.

Roger

Amit has a theory that large aftershocks will happen at 12 hour intervals
or 7 1/2 days after the main shock.

I looked at all mag 6+ quakes in the neic catalog (1973-2013)
The program read thru the file looking for a mag 7.5+ quake.
When it finds one it starts computing the hours until the next quake.
This continues until it finds a new main quake or until 8 days have passed.

Code:

HOURS AFTER THE MAIN QUAKE
| 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
|-----------------------------------------------------------------------
|86 30 22  7  5 10  6  6  7 12  9  6  7  6  8 12 13  1  6  2  5  6  6  8
| 8  6  3  3  5  5  2  5  5  6  2  4  3  5  2  3  4  3  5  3  2  4  6  1
| 1  4  1  2  5  3  5  1  3  5  2  5  2  1  1  1  5  1  2  0  2  6  2  1
| 1  1  2  6  2  2  1  2  5  3  7  1  3  3  3  1  5  0  1  4  1  4  2  4
| 4  2  1  6  2  1  2  3  3  3  5  7  4  2  0  5  1  1  2  1  5  5  4  6
| 6  2  4  1  5  3  3  3  3  0  4  4  7  2  3  1  7  2  2  5  5  4  3  1
| 1  5  1  3  6  5  2  5  4  0  2  5  1  1  1  2  2  5  0  2  1  2  3  6
| 6  3  4  3  0  1  0  3  4  1  4  2  2  2  4  2  4  0  2  3  3  4  1  2

Looking at the table you will see that there is no peak at 12 hours
and none at 7 1/2 days.

Roger

Quote:Roger
I can not make out anything from this. Are thse figures pertaining to a single quake of 7.5?
How many quakes of more than 7.5 have to plotted ?
Amit

No, the table is the cumulative results of all 7.5+ quakes from 1973-2013, looking for anything smaller up to 8 days later. There were several hundred of them I'll go back and count them if it is important.

Roger

New test; Amit has asked that people try to beat his success ratio.

I used as my data set the mag 6+ quakes from 1993-2013.
The odds on getting a hit in a 3 day window are 0.621.

I picked 600 random dates for that time, requiring each one to be at least 3 days away from any other selected date and checked to see if the corresponding window contained any quakes.

Results to follow.

Roger

(05-03-2014, 04:14 PM)Jim W. Wrote: I do believe your test failure rate is still rather HIGH!!!!. I asume that the reason for this is that there are still way to many unknowns to solve yet to have any form of relyable prediction other than there will always be earthquakes somewhwere at some time or other!