hi
Here are probable dates for major earthquakes in May 2019
4th May 2019......6.2
8th May 2019.....6.6
14th May 2019....6.5
18th May 2019....6.7
21/22nd May 2019.....6.8
here are the rules
1) window period is + or - one day
2) quakes predicted are 6.2+ and not 6+. Hence odds for 6.2+ can be taken for evaluation and not 6+ odds
21/22nd May 2019 means 12 Noon GMT on 21st to 12 Noon on 22nd
3) There are 5 dates predicted ,hence total 15 days window period in the Month. It means total 48.38% of the month is consumed in window period. Hence, if I get 48.38% of total quakes in the Month in my window period , it is exactly equal to odds. If more better than odds
Amit
Here are probable dates for major earthquakes in May 2019
4th May 2019......6.2
8th May 2019.....6.6
14th May 2019....6.5
18th May 2019....6.7
21/22nd May 2019.....6.8
here are the rules
1) window period is + or - one day
2) quakes predicted are 6.2+ and not 6+. Hence odds for 6.2+ can be taken for evaluation and not 6+ odds
21/22nd May 2019 means 12 Noon GMT on 21st to 12 Noon on 22nd
3) There are 5 dates predicted ,hence total 15 days window period in the Month. It means total 48.38% of the month is consumed in window period. Hence, if I get 48.38% of total quakes in the Month in my window period , it is exactly equal to odds. If more better than odds
Amit
44 comments:
hi
The probable dates are also posted on my web page
https://members.webs.com/MembersB/editAppPage.jsp?app=blog&pageID=208631614#blog/
Amit
Amit;
I must warn you that the chances of getting the mag right are very small.
You're better off predicting 6+ rather than 6.2
Roger
Amit;
Never mind; you said 6.2+ which is fine.
More work for me. Now I'll need a table of odds for all possible mags from 6.0+ to 9.9+
Any luck finding a list of all your predictions?
Roger
Hey terrific website! Does running a blog similar to this require a massive amount
work? I have virtually no knowledge of coding however I was hoping to start my own blog
soon. Anyway, should you have any ideas or techniques for new
blog owners please share. I know this is off topic nevertheless I simply had to ask.
Kudos!
Amit;
I finally finished fixing my copy of the NEIC catalog and ran all your predictions from 2006 to 2018 thru it. The prediction time span started 1 day before your first prediction and ended 1 day after the last prediction in 2018, 4422 days in all. That's 1472 possible 3 day windows. 895 of them had 1 or more quakes so the odds on a hit are 0.607
You made 451 predictions during the test period and 242 of them had 1 or more quakes but by chance you should have had 274 hits. Thus you are doing worse than chance.
Due to the large size of the prediction file and quakes I had to simplify things a bit so I could not consider partial day window boundaries or variable mag ranges. I considered all your windows to be 3 full days and all quakes to be 6.0+
You could claim that this was responsible for your low score but you would need to prove it.
I'll be happy to supply copies of the prediction list if you wish to verify it's accuracy.
Roger
Roger
Here is some actual calculation and results
Year 2011
Predicted (p)....38
Actual hit(h)......27
Year 2015
P....39
h....29
2016
P....43
h......30
2017
P......40
h......21
2018
P...41
hit...26
Total results
Predicted....201
Hit..........133
Percentage..66.17
Will do rest of calculation as time permits
Amit
Amit
Amit;
You seem to be counting predictions and those containing quakes.
What mag level are you using?
A few selected years prove nothing.
Roger
Amit;
If I set the minimum mag to 6.2 the results are:
There were 451 predictions and 180 hits.
There were 4421.9993 days in the test interval.
There were 1474 windows and 669 with quakes.
Odds on a hit are .45386703
Expected number of hits is 204.7
Still below chance.
Roger
Roger
I count 6+. because earlier I have already mentioned this on my blog
Yes ,these are my predicted dates and actual hit in window period
I will be giving all years data one by one , because I have to do physically.
However no chances of error ,mistake or bug
Amit
I am no longer certain the place you are getting your info,
but good topic. I needs to spend a while finding out much more
or working out more. Thank you for excellent information I was looking for this
info for my mission.
This information is worth everyone's attention. When can I find out more?
Hi
Pl note 6.2 ( modified by USGS from 6.1) on 3rd ,a 6.2 quake prediction
Here is the solution for the recent event.
May 3, 2019, SOLOMON ISLANDS, MW=6.2
Howard Koss
CENTROID-MOMENT-TENSOR SOLUTION
GCMT EVENT: C201905030725A
DATA: II LD IU DK CU MN G IC GE
GG KP
L.P.BODY WAVES:165S, 382C, T= 40
MANTLE WAVES: 134S, 213C, T=125
SURFACE WAVES: 168S, 413C, T= 50
TIMESTAMP: Q-20190503073424
CENTROID LOCATION:
ORIGIN TIME: 07:25:34.5 0.1
LAT: 6.89S 0.00;LON:160.17E 0.00
DEP: 29.1 0.2;TRIANG HDUR: 3.1
MOMENT TENSOR: SCALE 10**25 D-CM
RR=-1.300 0.012; TT= 1.730 0.011
PP=-0.432 0.012; RT= 0.921 0.024
RP= 0.726 0.025; TP=-1.610 0.010
PRINCIPAL AXES:
1.(T) VAL= 2.649;PLG= 7;AZM= 26
2.(N) -0.259; 43; 289
3.(P) -2.392; 46; 124
BEST DBLE.COUPLE:M0= 2.52*10**25
NP1: STRIKE=154;DIP=53;SLIP= -31
NP2: STRIKE=263;DIP=65;SLIP=-139
###########
--############# T #
---############## ###
-----######################
-----########################
------#########################
-------######------------######
-------#------------------------#
---#####-------------------------
-########------------------------
##########------------ --------
##########----------- P -------
###########---------- -------
###########------------------
############---------------
############-----------
#############------
###########
These are in fact wonderful ideas in concerning blogging.
You have touched some fastidious factors here. Any way keep up wrinting.
I love your blog.. very nice colors & theme.
Did you make this website yourself or did you hire someone to do
it for you? Plz reply as I'm looking to create my own blog and would like to know
where u got this from. many thanks
Roger
received your e mail regarding results of my predictions. However I could not make out any conclusion from the table forwarded by you
ca you please post results in concise form over here
Amit
roger
Here is one more year -2014- prediction. This year not good
total dates predicted dis 41 and hit in window period is 27(6+)
thus ratio is 27/41 ie 65.8%
which is still above average
If anyone wants month wise data , I can post here.
For Feb 2014 prediction was not done ,This is perhaps the first and last month for which I have posted the dates
Amit
Amit;
Finally finished the analysis of all your predictions.
Results are:
There were 456 predictions and 245 hits.
There were 4421 days in the test interval.
There were 1474 windows and 895 with quakes.
Odds on a hit are 0.607
Expected number of hits is 276.9
Below chance.
Roger
Hi Total 160 hits in years
2011,12,14,15,16,17,18
Still 2006 to 2010 remains
Amit
Amit;
If you come up with a different number of hits we will have to go over them to see where the error is.
A hit is one or more quakes of the right size (6+) within a predicted time window.
Roger
hi
Here is my worst prediction year 2013
Total date predicted are only 37
hits in window are only 20 percentage hit is 54%
However there is reason for this
My wife was operated thrice for her abdominal problems in the year 2013 and I was in Hospital for days together with her
anyway ,this is not the excuse I know
AMit
Amit;
I've tested all your predictions for 6+, 7+ and 8+ quakes and none of them are better than chance.
Your method needs revision or rejection.
Roger
Amit;
You should move my evaluation posts to your latest group
Roger
Amit;
You may not realize the power of this program.
Using it, I can look at any part of the NEIC data, from 1973 to 2018.
That's over 700,000 quakes of all sizes.
I have a trimmed down copy of the entire catalog, containing just the
quake date, time, location, depth and magnitude
I can look at any subset of your predictions from 2006 to 2018, from a
single year to the entire list.
You name it, I can probably do it.
Roger
roger
such table is seen fwd by you. i could not make out anything from this
It is a long table of some 415 rows
Amit
19 11 2006 0 0 IST 6.5
11 12 2006 12 0 IST 7.2
18 12 2006 0 0 IST 6.9
24 12 2006 0 0 IST 6.4
Amit;
Those are your predictions, exactly as you made them.
Day Month year hour minute Time zone mag.
I ignored time in evaluating them.
Roger
Hi
I am aware,when I write this post ,that, this can not be treated as a hit prediction . A 7.2 quake at PNG on 6th 21.20 hrs GMT
However ,watch inaccuracy of 2 hrs 40 minutes(earlier)
Nevermind ,watch till window period ends
Amit
Amit;
I can't handle these noon to noon predictions; it would double the program time because each prediction would have to be checked to see which test to perform.
Roger
Roger
Ok
No problem
You can take the first date in all such predictions
Amit
Hi
A 6.3 quake at at Japan on 9th ,a window hit prediction
Amit
Hi
https://m.emsc.eu/earthquake/latest.php?min_mag=6&max_mag=n/a&date=n/a&euromed=World
emsc says quake is 6.4
USGS portal says ...6.1
an e mail from USGS says .6.3
You decide yourself how much accurate the seismologist are
Amit
Amit;
What difference does it make? Mag is an average from all the data you have.
That changes with time as more come in.
Roger
Roger
Here is the point
As this time I have stated 6.2+ quakes and not 6+, chances are you declaring this as a miss prediction,stating it is 6.1 and not 6.2
I wonder when it comes USGS data discrepancy you always defend USGS.
Even after the quake has occurred ,data is not consistent ,while you may not me to treat 6.1 as 6.2 quake
USGS has a whole battery of staff , equipments and money along with super computers and seismometers, while I work part time ,single handed ,without money ,without staff and any software.
This is pure injustice
Amit
Roger
But after 24 hrs of quake ,there should be one figure on all data sources
Amit
Roger
The interesting thing is ...
You are comparing my prediction with the database ,and declaring it as a miss by 0.1 or 0.3 mag, is in itself is not that accurate
Amit
Hi
If anyone has problem posting comments on this blog ,pl contact
a.amitjdave@rediffmail.com
Amit
Hi
My reader friends are requested to post as a trial post as to whether one can upload a genuine related post, I doubt ,some hackers have made mischief.
On my setting search ,it appears some viewers are redirected from it to porn sites through or by my blog
People are earnestly request to stop such things in larger interest and for the sake of science and ethics
Please treat this as a humble request from an old ,lone , passionate man
Thanks
Amit
Amit;
What you need to do is examine all 6+ quakes for whatever you think is significant; moon phase, moon distance, etc. etc.
Then see how many fit your theory.
I have examined all your predictions to see if they captured quakes. Some do, but not enough of them to show significance.
Your theory is wrong. You need to modify it until it can be relied on or abandon it entirely.
Roger
Hi
For all those who strongly believe that major Earthquakes can not be predicted,here is a 7.7 date hit prediction
7.7 at PNG on 14th May 2019
Amit
Amit;
That sounds impressive but it's actually deceptive, leading one to think you are always correct.
That is of course, not true.
I know; I have computed your long-term average and it's below chance.
You should quit until you can do better.
Roger
Amit;
I can help with computer programs if you can think of anything that can be tested that way.
No charge; I do it to keep my brain from rotting.
Roger
Roger
A 6.2( as predicted) Earthquake at new Caledonia on 19th early morning .
Watch the 6.2+ accuracy.
Amit
Hi
One more hit prediction
6.1 quake at Alaska on 23rd May 2019
Amit
amit;
mag 8.0 today. You missed it
WHY????
Roger
Amit;
You had a good month for a change; all 5 of your windows were correct.
However, only 7 of 12 quakes were in those windows.
So, just chance.
Roger
Post a Comment