Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

[metalman]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

And how does Kevin Bacon fit into all of this?

[Fiat Currency]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

back to the point... how do we explain the djia ending at the same price the same day and only on that day?

theories?

C O I N C I D E N C E

coincidence.jpg

[sunskyfan]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

I took a look at the DOW closes. Repeatable numbers are common as is numbers that have never been reached. Coming up with real probability on such a thing is not easy. I decided the odds were about 1 in 8000. Not incredible but not easily to dismiss considering the context.

[metalman]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

I took a look at the DOW closes. Repeatable numbers are common as is numbers that have never been reached. Coming up with real probability on such a thing is not easy. I decided the odds were about 1 in 8000. Not incredible but not easily to dismiss considering the context.

thx, suskyfan. you mean the same number on a one year anniversary date are common?

[sunskyfan]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

No ... just that numbers being repeated on different days are common within the sample set.

[MulaMan]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

Well that is a great theory as to why wall-street runs the country but I still like my theory better and so blame Ronald Reagan for starting this entire rule the world by printing money and sell off America thing in order to win the cold war.

[ThePythonicCow]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

No ... just that numbers being repeated on different days are common within the sample set.

This sounds like it could be a variant of the Birthday Paradox.

[metalman]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

No ... just that numbers being repeated on different days are common within the sample set.

obvious fact is obvious

my script takes the adj close of the djia for feb 1, 1929 then looks at the adj close for feb 1 every year for 60 years. then does the same for feb 2, 1929 and every day of every year after that... ave. 200 days per yr x 60 years for 60 years.

the chances are 1 in several million, i figure. inflation makes repeats 10 yrs years out highly unlikely. when will we see dow 340 again? ;)

the most likely time periods fo adj close to repeat on the anniversary were during long bear markets... 1930s... 1960s... and since 20001.

found only 1 occurrance... 09-11-01 & 09-11-09.

coincindentally, i believe there was a terrorst attack that day, too.

[Diarmuid]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

This sounds like it could be a variant of the Birthday Paradox.

Not sure TPC

the equation you see in the birthday paradox is

It is using just a static subset of 365. In the DOW case I suppose the chance of two days closing at the same number, one must factor in inflation, i.e the dow closing today at 1965 figures is nearly unthinkable, so there is a time factor i.e as time increases between data points the likely hood of the two matching becomes less and less likely, baring some extreme event. Because we have had an extreme event last year I suppose the chances of correlation between a day in 2001 and 2009 may have become more likely, but I think any attempt to find the statistical probability would have to take that into account and also the fact that the decreasing probability of two data points matching with an increase in time?

Maybe way to obvioius and you had already thought about it in these terms or I have skewed my logic, thinking out loud is all.

Thoughts :confused:

Edit: MM seems you and I had the same thought at the same time :-)

[jckb98]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

Hey Metalman,

I respect your views on many other issues you know, but you might be spooking out some iTulipers on this DJIA thing...and your reputation is making others to group-think instead of fact check, which in my opinion, is really BAD for the forum.

If you look back in history, the stock market was not traded on September 11, 2001. The closing market number you have, was from Sept 10th, the day before. The market re-opened on 17th of Sept, 2001.

So technically, if you wanted to compare non-open market dates (never mind the 0.1 difference that exist but you said you rounded off. However, this is exactly what EJ has told us to be careful about, DO NOT MAKE THE DATA FIT YOUR THEORY and DO NOT THROW OUT NON-THEORY FITTING DATA, and instead DO investigate them), you might as well compare Sept 12th-13th 2001, vs Sept 12th-13th 2009 all having the same DJIA closing numbers, since all are non operating days (2001 = attack, 2009 = weekend).

Lets end this thread. Your reputation is much too important than to be dragged through the mud over this, 0.1 margin coincidence.

[ThePythonicCow]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

Not sure TPC

the equation you see in the birthday paradox is

Well, certainly the specific equation of the Birthday Paradox doesn't apply here, nor does any ordinary variation of that equation using different factors.

But there is a psychological side to the Birthay Paradox as well. It is a side seen in the fascination with numerology and other such strange coincidences. It is a side seen on first dates, when people delight in the strange coicidences they find between themselves.

I do not know of any useful mathematical model to calculate probabilities in this space, because a critical parameter cannot be estimated, even within orders of magnitude. That parameter is how many such coincidences, if they occurred and were noticed, would seem strangely interesting. The number of such is humongous, mind boggling stupendous. Our brains seem wired to filter for such and notice them.

[Diarmuid]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

That parameter is how many such coincidences, if they occurred and were noticed, would seem strangely interesting. The number of such is humongous, mind boggling stupendous. Our brains seem wired to filter for such and notice them.

LOL - bravo TPC

[metalman]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

Hey Metalman,

I respect your views on many other issues you know, but you might be spooking out some iTulipers on this DJIA thing...and your reputation is making others to group-think instead of fact check, which in my opinion, is really BAD for the forum.

If you look back in history, the stock market was not traded on September 11, 2001. The closing market number you have, was from Sept 10th, the day before. The market re-opened on 17th of Sept, 2001.

So technically, if you wanted to compare non-open market dates (never mind the 0.1 difference that exist but you said you rounded off. However, this is exactly what EJ has told us to be careful about, DO NOT MAKE THE DATA FIT YOUR THEORY and DO NOT THROW OUT NON-THEORY FITTING DATA, and instead DO investigate them), you might as well compare Sept 12th-13th 2001, vs Sept 12th-13th 2009 all having the same DJIA closing numbers, since all are non operating days (2001 = attack, 2009 = weekend).

Lets end this thread. Your reputation is much too important than to be dragged through the mud over this, 0.1 margin coincidence.

you are correct. got so busy trying to find anniversary dates matches i forgot to check to see if the premise is wrong... that 091101 did not match 091101.

could have guessed given the, er, lack of response from my pals here.




just goes to show... even a skeptic can get carried away.

thx for pointing out the error. case closed!

[sunskyfan]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

Just answering your question ... sum facts

20,237 days of closes ...

min 190 and max is 14164

between that range 8701 values have never occurred

2437 values have occurred at least twice with the maximum number of recurrences being 85 at Dow 180 way back when. Above Dow 1000 the max number of recurrences is 12.

Looking at all the numbers: The maximum stretch between same closes is at Dow 320 and was 6332 days. The average maximum separation between same closes is actually 1142 closes. The median maximum separation between same closes is 458 closes.

Looking at just the Dow closes above 1000: the average maximum separation is 491 closes and the median maximum separation is 207 closes.

So it looks like this is not so hard to do but if you look at that between 1000 and 14000 a particular close value only has a 52% chance of even happening and only 22% of those numbers happen at least twice leaves a probability of 11% of a number between 1000 and 14000 ever happening twice.

Between 1000 and 14000 ...

All this just to get a feel .... the best WAG is that there 1556 closes with repeat values at 6886 values that happened. Choosing a value 1 out 6886 and the fact that there is only a 16% chance of it being repeated means .16/6886 or a 1 in 43037 chance of choosing a specific number again out of the group of available values.

Not enough to prove "fix" perhaps but low enough not to call somebody crazy if they wondered what was going on.

[tombat1913]

Re: Spooky: DJIA Closed at 9,605 on 9/11/01 & 9/11/09

On a similar note.

On Wednesday, Sept. 11, 2002 — 9/11/02 — the New York State lottery numbers were 911, an eerie coincidence.

Their analysis of the odds being 1/1000 is true for any given day, the odds of it happening exactly on the anniversary is a whole different story. Hey, reporters aren't stisticians.