The "Hindenburg Omen" now statistically confirmed, and we just had another one.

For some reason this thread reminds me of two headlines I have never seen:
1. "Psychic Wins Lottery Again" and
2. "Top Ten Richest People In The World Are All Economists"
 
paiute said:
For some reason this thread reminds me of two headlines I have never seen:
1. "Psychic Wins Lottery Again" and
2. "Top Ten Richest People In The World Are All Economists"
Click to expand...

3: How I discovered the "Hindenburg Omen" and made a billion dollars.
 
So, one condition of successful prediction is that the value of the entire NYSE will be reduced by at least 2% within 17 weeks of the date the linked article was published?
 
31 of 31 for the researchers is a good track record. If you were told by such people that 7 times out of 8 it rained after there was thunder, would you take your umbrella with you to work the next time it thundered or not?
 
Comrade Quixote said:
Predictive formulae are only useful if they work forwards, i.e. if you can predict what happens, not if you alter them after they have failed to do so so that they would have predicted the future.
Click to expand...
Ditto on this.

It also helps to have a model that drive the addition (or removal) of conditions, before we needed to realize they were necessary (or not) from the data.

In other words, the Hindenburg Omen is saying this: "Well, the data fits this pattern, so far, so let's roll with that. And, if it breaks will alter the 'pattern' to one that would then fit the data, afterwards."

It would be more powerful to say something like this: "Well, here is a model for how stock prices will tend to fluctuate, based on natural tendencies we discovered in market evolution. If this breaks either the model was wrong, or selection pressures had to have radically changed in a short amount of time, which is unlikely."

That second, more powerful, type of model does exist in a very, very vague manner, that is only useful for studying general trends, and not for buyers and sellers. (Well, perhaps a certain class of long-term holders.) If one could develop a more precise model, that is actually useful to buyers and sellers, THAT would be amazing!
 
Muldur said:
31 of 31 for the researchers is a good track record. If you were told by such people that 7 times out of 8 it rained after there was thunder, would you take your umbrella with you to work the next time it thundered or not?
Click to expand...

Depends who told me, and how credible they are.
 
Plus one has to recognize that the stock market goes up and down on its own daily and this is normal. Just bringing up the CNN Index shows me massive variance just this week. Going to a five year view of the chart shows the fractal nature of this pattern of ups and downs. It goes up, it comes down a little, it goes up again. Big fat hairy deal.

Anyone can read that pattern and see that it has been swinging up lately and is due for another down. That's like knowing that when you push a swing it will come back. When stocks get to a certain height the wise investor sells. Enough wise investors start selling and the whole thing starts coming back down. This isn't hard science, it's sociology!

Edit: I would note that any model that was widely known about would effect the very thing it was trying to model and thus render itself no longer viable. The stock market is about how much people think something is worth, not its real value. Just making people think that there is a coming crash can cause their behaviors to change.
 
Last edited:
Muldur said:
And of course you completely ignore that that is a MINIMUM value, and that there is a high probability (67%) of a 5% or greater drop and lesser but still significant probabilities of even steeper losses.
Click to expand...

You need some sort of a blimp-related omen to tell you that there's a significant chance that the market will fluctuate during the next four months? :boggled:
 
Well... technically speaking, the Hindenburg was not a blimp.
 
From this blog:

What's a Hindenburg Omen? There are two decent answers. The first is the technical definition: The Omen is triggered when: (1) more than 2.2% of stocks on the NYSE are at 52-week highs AND more than 2.2% are at 52-week lows, (2) the 50-day moving average is trending higher, (3) the McClellan Oscillator is negative, and (4) new 52-week highs don't exceed new lows by more than 100%.

When that happens twice within 36 trading days the Omen is triggered. All four conditions were met on April 15th, metaphorically lifting an enormous bag of Nazi-sponsored hydrogen over the stock market. To devotees of the Hindenburg Omen Friday's close was a static electricity spark over Lakehurst New Jersey. Now we can only sit back and watch the stock market burn. Oh, the humanity indeed.
An alternative explanation is that the Hindenburg Omen is BS. Supporting the latter argument is the fact that the inventor of the Omen received wide attention when he sold all his stocks in the last week of August 2010, immediately prior to the 25% rally over the next 6 months.
Click to expand...

:dl:

31 for 31 indeed.
 
You must log in or register to reply here.

Back
Top Bottom