5 ms·

Hindenburg Omen (Occurred Twice this Month)

40 points by prosa 16y ago

ojbyrne 16y ago
The consensus here was that this was drivel the first time it appeared (2 or 3 days ago).
- NathanKP 16y ago
  I would tend to agree, although the Hindenburg omen might easily be explained by publicity like this submission on HN. Investors read about the Hindenburg omen and the fact that it has occurred twice this month. They assume that a crash is imminent, so they start selling stocks, which causes the market to collapse. In this way what is drivel could be made reality because too many people don't believe it to be drivel. There are some really interesting analysis of stock market dynamics such as this in the book "The Wisdom of Crowds" by James Surowiecki
msy 16y ago
That's a hilariously small sample size for a 'model'.
raquo 16y ago
Not that I'm a finance expert, but I'm uncomfortable with an indicator the parameters of which were optimized by backtesting half a century ago. On the other hand, a confirmed Hindenburg Omen signal occurred on Jun 16th 2008, which would be a timely warning for the following events. On yet another hand, no word on how many such confirmed signals were false alarms historically.
- sdfx 16y ago
  Doesn't it say in the article that in the past, 24% of the Hindenburg Omen were followed by a stock market crash? From the article: Looking back at historical data, the probability of a move greater than 5% to the downside after a confirmed Hindenburg Omen was 77%, and usually takes place within the next forty-days. The probability of a panic sellout was 41% and the probability of a major stock market crash was 24%. While the past performance is quite compelling, the criteria seem to be extremely arbitrary.
  - raquo 16y ago
    The "Conclusions" section of this wiki page is rather uninformative... See http://www.safehaven.com/article/3880/the-past-performance-of-the-hindenburg-omen-stock-market-crash-signals-1985-2005 http://www.safehaven.com/article/3880/the-past-performance-o... (from external links) There's a table with all 22 confirmed signal clusters 1985-2005 and the following "crashes" along with their delays. Although the phrase "25% of these signals were followed by a market crash" is pretty impressive, when you look at the actual definitions of crash (>15% drop), and the delays (up to 4 months) it doesn't look as exciting anymore.
Tichy 16y ago
Funny that probably a lot of people make a living thinking in such a way, forming complicated theories of superstition.
- davidw 16y ago
  It sounds like the beginning to one of those 80ies barbarian movies (I'm thinking of The Beastmaster, which is a must-see): "On the 10th day, the Omen foretold the coming Prophecy that the Evil Emperor would be slain by a Child born under the Star of the Omen"
  - Tichy 16y ago
    Would be fun to see a SciFi movie based on this idea :-)
ANH 16y ago
I've never seen anything that makes me think 'Technical Analysis' of stock market indicators is anything more than a pseudoscience. It's fun and spooky, but is it any better than throwing darts?
- akkartik 16y ago
  In an ideal world, technical analysis tries to model psychological effects that are statistically significant in a population of humans. Don't buy just because you see a W, or crap like that. But if you understand why a W is an interesting pattern, and what it's based on, it can be a valuable signal to digest. I don't know enough to trade purely based on technicals, and I'm not sure anybody should. But when I want to buy a stock I wait until it's oversold (http://clearstation.etrade.com/cgi-bin/details?Symbol=goog http://clearstation.etrade.com/cgi-bin/details?Symbol=goog) That usually saves a few bucks.
- Keyframe 16y ago
  It kind of works because everyone else thinks it works. Self fulfilling prophecy. It's like looking at traffic lights and thinking traffic should move the way they dictate it - it works because, almost, everyone participating in traffic follows the same signs.
- btilly 16y ago
  No, it is generally complete BS. It survives on human gullibility, and the observation by stock brokers that promoting technical analysis is good for trading volume. A good book that addresses this A Random Walk Down Wall St. That said the particular "omen" they are talking about does have some merit. Normally there is a broad correlation between different stocks, so you don't typically get both lots of highs and lows at the same time. (This correlation is why the DOW, which just tracks a small number of indicator stocks, is correlated with much broader sets of stocks.) But you can get lots of highs and lows at the same time if the market is very volatile. Volatile markets sometimes go down very sharply. Of course causation goes the other way. Fears that the market could drop cause volatility, and sometimes those fears are proven right. That said we can and do measure market volatility directly. (For instance every options pricing model does that.) And that gives a much more fine-tuned predictor of risk than this "omen".
CWuestefeld 16y ago
This is right up there with the similarities between Kennedy and Lincoln.
jonknee 16y ago
Sometimes humans too good at seeing "patterns" for our own good.
- jokermatt999 16y ago
  There's a word for that: http://en.wikipedia.org/wiki/Paredolia http://en.wikipedia.org/wiki/Paredolia
jacoblyles 16y ago
OMG, let's go looking for Ms and Ws in stock price data
c1sc0 16y ago
Let's assume the big crash will happen before the end of this year: what do you do? As an angel? As an entrepreneur trying to raise money? As a funded startup with a limited runway?

40 points by prosa 16y ago

ojbyrne 16y ago

The consensus here was that this was drivel the first time it appeared (2 or 3 days ago).

NathanKP 16y ago
I would tend to agree, although the Hindenburg omen might easily be explained by publicity like this submission on HN. Investors read about the Hindenburg omen and the fact that it has occurred twice this month. They assume that a crash is imminent, so they start selling stocks, which causes the market to collapse. In this way what is drivel could be made reality because too many people don't believe it to be drivel. There are some really interesting analysis of stock market dynamics such as this in the book "The Wisdom of Crowds" by James Surowiecki

msy 16y ago

That's a hilariously small sample size for a 'model'.

raquo 16y ago

Not that I'm a finance expert, but I'm uncomfortable with an indicator the parameters of which were optimized by backtesting half a century ago. On the other hand, a confirmed Hindenburg Omen signal occurred on Jun 16th 2008, which would be a timely warning for the following events. On yet another hand, no word on how many such confirmed signals were false alarms historically.

sdfx 16y ago
Doesn't it say in the article that in the past, 24% of the Hindenburg Omen were followed by a stock market crash? From the article: Looking back at historical data, the probability of a move greater than 5% to the downside after a confirmed Hindenburg Omen was 77%, and usually takes place within the next forty-days. The probability of a panic sellout was 41% and the probability of a major stock market crash was 24%. While the past performance is quite compelling, the criteria seem to be extremely arbitrary.
- raquo 16y ago
  The "Conclusions" section of this wiki page is rather uninformative... See http://www.safehaven.com/article/3880/the-past-performance-of-the-hindenburg-omen-stock-market-crash-signals-1985-2005 http://www.safehaven.com/article/3880/the-past-performance-o... (from external links) There's a table with all 22 confirmed signal clusters 1985-2005 and the following "crashes" along with their delays. Although the phrase "25% of these signals were followed by a market crash" is pretty impressive, when you look at the actual definitions of crash (>15% drop), and the delays (up to 4 months) it doesn't look as exciting anymore.

Tichy 16y ago

Funny that probably a lot of people make a living thinking in such a way, forming complicated theories of superstition.

davidw 16y ago
It sounds like the beginning to one of those 80ies barbarian movies (I'm thinking of The Beastmaster, which is a must-see): "On the 10th day, the Omen foretold the coming Prophecy that the Evil Emperor would be slain by a Child born under the Star of the Omen"
- Tichy 16y ago
  Would be fun to see a SciFi movie based on this idea :-)

ANH 16y ago

I've never seen anything that makes me think 'Technical Analysis' of stock market indicators is anything more than a pseudoscience. It's fun and spooky, but is it any better than throwing darts?

akkartik 16y ago
In an ideal world, technical analysis tries to model psychological effects that are statistically significant in a population of humans. Don't buy just because you see a W, or crap like that. But if you understand why a W is an interesting pattern, and what it's based on, it can be a valuable signal to digest. I don't know enough to trade purely based on technicals, and I'm not sure anybody should. But when I want to buy a stock I wait until it's oversold (http://clearstation.etrade.com/cgi-bin/details?Symbol=goog http://clearstation.etrade.com/cgi-bin/details?Symbol=goog) That usually saves a few bucks.
Keyframe 16y ago
It kind of works because everyone else thinks it works. Self fulfilling prophecy. It's like looking at traffic lights and thinking traffic should move the way they dictate it - it works because, almost, everyone participating in traffic follows the same signs.
btilly 16y ago
No, it is generally complete BS. It survives on human gullibility, and the observation by stock brokers that promoting technical analysis is good for trading volume. A good book that addresses this A Random Walk Down Wall St. That said the particular "omen" they are talking about does have some merit. Normally there is a broad correlation between different stocks, so you don't typically get both lots of highs and lows at the same time. (This correlation is why the DOW, which just tracks a small number of indicator stocks, is correlated with much broader sets of stocks.) But you can get lots of highs and lows at the same time if the market is very volatile. Volatile markets sometimes go down very sharply. Of course causation goes the other way. Fears that the market could drop cause volatility, and sometimes those fears are proven right. That said we can and do measure market volatility directly. (For instance every options pricing model does that.) And that gives a much more fine-tuned predictor of risk than this "omen".

CWuestefeld 16y ago

This is right up there with the similarities between Kennedy and Lincoln.

jonknee 16y ago

Sometimes humans too good at seeing "patterns" for our own good.

jokermatt999 16y ago
There's a word for that: http://en.wikipedia.org/wiki/Paredolia http://en.wikipedia.org/wiki/Paredolia

jacoblyles 16y ago

OMG, let's go looking for Ms and Ws in stock price data

c1sc0 16y ago

Let's assume the big crash will happen before the end of this year: what do you do? As an angel? As an entrepreneur trying to raise money? As a funded startup with a limited runway?