Roger
Now ,at least to prove your point that my predictions are below average,you should give some dates
with equal numbers of date hits.
If you can not do that,it means there is something wrong with your average calculation methods( basically due to clusters)
One which is theoretically correct should prove practically.
Every time you ask me to prove the date hits. Now ,after series of date hits, onus is on you to prove practically, that ,you or any body can give equal number of date hits
Amit
10 comments:
Ah, an interesting idea Amit. Suppose we all have a shot at forecasting the month of July. Perhaps having a posting deadline of 12UTC, June 30, for anyone wishing to contibute. Then, we may have a statistical discussion as to the results.
Steve
Amit;
Such a test would prove nothing except that you know nothing about statistics.
ANY result can be obtained by chance in a small sample. You might win, I might win, it doesn't matter.
What does matter is the result of a long series of trials.
Roger
Roger
Have you ever tried to evalute the wheather forecast statistically?
Are the forecast below average or above average?
In fact every day,wheather department gives a forcast valid for next 72 Hrs.Thus, the window of forecast never closes.We believe
the forecast and Government spends fortunes in insatlling equaipments
regards
Amit
Amit;
Weather forecasts are a bad analogy. The weather results from a large number of observable variables. The problem is getting enough information and processing it fast enough.
Earthquakes result from forces which cannot be directly observed so the quakes appear random.
Roger
OK Amit,
here's my toss of the coin for July. Each date with a "3" is a forecast of a least 1 M 5.7+ EQ.
Each blank date is a forecast of 0 M 5.7+ EQ.
Also would like to go on record as forecasting a M 8.0+ EQ on 7/26 or in the period of days 7/25 thru 7/30
Also, Amit as per this discussion I have already posted a randomly chosen forecast to compare with my astrological forecast on my web AstroWeather.com
7/1 3
7/2 3
7/3
7/4
7/5
7/6
7/7 3
7/8 3
7/9 3
7/10 3
7/11 3
7/12 3
7/13 3
7/14
7/15
7/16
7/17
7/18
7/19
7/20
7/21 3
7/22 3
7/23
7/24
7/25 3
7/26 3
7/27 3
7/28 3
7/29 3
7/30 3
7/31
Roger,
my June forecast is now history, and my score is 22 successful results out of 29 trials, for a total score of 76% correct, given a 1 day 'window'.
How does your prediction evaluation formula describe this result?
Steve DeLapp
Sdelapp
Yeas. Yuo are right. July end 2010 will see some of the major quakes. please see my blog dt 10th January 2010 where the dates were already predicted well in advance
regards
Amit
sdelapp;
I have tested your June predictions and find you have 21 out of 30 predictions correct.
This is significant at the 97.8% level which does not meet my 99% test level but is still the best I have seen so far.
However, a 30 day test is far too small to mean anything. You need about a hundred predictions to start feeling confident.
Roger
hi
For deails of theory and predictd dates (calender) please se emy web page
http://earthquakeprediction.webs.com/
Amit
On Aug 15 i predicted a huge earthquake i live in the dc area and 8 days later a big quake for dc Hit below is the email i sent i took out the part of the name and his email for privacy
RE: i predict a huge earthquake this week
X
Inbox
X
Reply
JD to me
show details Aug 15
Huh?
---------------------------------------------------------------------
J.D.
Rockville, MD 20857
(301)
jfritz
From: David Fairbanks [mailto:dcfairbanks@gmail.com]
Sent: Monday, August 15, 2011 8:28 AM
To: J. D.
Subject: i predict a huge earthquake this week
Post a Comment