3 ms·

What exactly was the new information here? It was a completely fabricated article made to look like it came from an industry source. The article has no bearing

What exactly was the new information here? It was a completely fabricated article made to look like it came from an industry source. The article has no bearing on Twitter, yet the shares traded up 8% immediately after it came out and fell back down as it was realized the article was farce. If any new information was created, it's that whoever is trading Twitter stock (and by extension, most other stocks) trades on false information at least some of the time. My strong hypothesis is that most trades are based on false information most of the time.

kasey_junk 11y ago
New piece of information A - "Twitter might have got a bid buyout" New piece of information B - "That last story was completely fabricated". I'm not EMH believer but to claim that any version but the strongest (which is known to not apply) was implicated by this story is just misrepresenting what the EMH says.
- jsprogrammer 11y ago
  A (which is arguably not real information) occurred simultaneously with B (actually, I might argue B occurred before A) as the fact that the story was fabricated was evident in the previously unknown and unauthenticated URL which was used to retrieve the "information".
  - kasey_junk 11y ago
    The fact that the information is not real is immaterial to the EMH in any of its forms. The strongest form (which is not applicable here) would indicate that the price jump shouldn't have happened if it was real the people who knew about that information first, could use it to make money and thereby naturally adjust the price of the equity so that there was no "jump". Given that the strong form CANNOT exist in the US equities market, you get situations where new information is allowed to cause price movements as long as it happens quickly and in an unbiased fashion, which this did. Think about it this way, the first piece of information is "Somebody says twitter is being bought out". That is information that should change the price, and that is the headline. "Anyone who says twitter being bought out is full of crap" should also change the price based on the 'reputation' of the people making the statements. Regardless of order those price movements are central to the EMH, not contradictory, and they did in the order we would expect in such a way as to back up the weaker forms of the EMH. So whether you buy the EMH, think even if it exists it doesn't matter, or think it is complete nonsense. This behavior is not evidence to back up your stance. If you think it is, you don't understand the EMH.
    - jsprogrammer 11y ago
      >The fact that the information is not real is immaterial to the EMH in any of its forms. Hmm, really? I guess I really don't understand the EMH then. Can you give me a formulation of EMH that doesn't depend on information? I tend to think that the EMH is a tautology and makes no real claims. It's a novelty. Back to our argument though, you're still disregarding the sequence of events. The reputation of the information precedes the information itself. It was possible to know to disregard the content before actually seeing the content. You keep claiming that the content came first, when that is strictly not the case. I will accept that many traders treat any text they come across as gospel and blindly follow it. I'm sure there's a continuum to someone who first verifies all information they depend on in a trade.
      - kasey_junk 11y ago
        I'm not making myself clear. The EMH does require information for any of its formulations. It doesn't require the market to magically determine what is "real" vs "fake" information and disregard "fake" information. In fact quite the opposite. The EMH is based on the idea that lots of people react to incorrect information all the time, but that just as many are working on correct information such that the aggregate is efficient. That is precisely what happened here. Some people reacted incorrectly to new information (ie they believed it) others reacted correctly (ie they saw it was nonsense) and the market in aggregate self corrected. I'm not sure that the correction falls into the semi-strong form as I'm not sure what the boundaries for speed of market reaction are. I think it is smart to be skeptical of the EMH, lots of people are for very good reasons, but this event is not evidence one way or the other.
  - prostoalex 11y ago
    A was simultaneous with B only in retrospect. It requires one basic skill (reading) to comprehend A, it requires a variety of skills (understanding of domain name system, technical expertise, knowledge on how to do lookups of a domain owner) to comprehend B. If article A had a large banner saying THIS STORY IS FAKE, you could argue A and B were simultaneous.
    - jsprogrammer 11y ago
      One must have the domain name (and in fact, the entire URL) before one can access the content (A). So, B actually came before A (assuming the article was only accessed over the Web [eg. not also handed out in print on the floor of a prop trade desk]).

kasey_junk 11y ago

New piece of information A - "Twitter might have got a bid buyout" New piece of information B - "That last story was completely fabricated". I'm not EMH believer but to claim that any version but the strongest (which is known to not apply) was implicated by this story is just misrepresenting what the EMH says.

jsprogrammer 11y ago
A (which is arguably not real information) occurred simultaneously with B (actually, I might argue B occurred before A) as the fact that the story was fabricated was evident in the previously unknown and unauthenticated URL which was used to retrieve the "information".
- kasey_junk 11y ago
  The fact that the information is not real is immaterial to the EMH in any of its forms. The strongest form (which is not applicable here) would indicate that the price jump shouldn't have happened if it was real the people who knew about that information first, could use it to make money and thereby naturally adjust the price of the equity so that there was no "jump". Given that the strong form CANNOT exist in the US equities market, you get situations where new information is allowed to cause price movements as long as it happens quickly and in an unbiased fashion, which this did. Think about it this way, the first piece of information is "Somebody says twitter is being bought out". That is information that should change the price, and that is the headline. "Anyone who says twitter being bought out is full of crap" should also change the price based on the 'reputation' of the people making the statements. Regardless of order those price movements are central to the EMH, not contradictory, and they did in the order we would expect in such a way as to back up the weaker forms of the EMH. So whether you buy the EMH, think even if it exists it doesn't matter, or think it is complete nonsense. This behavior is not evidence to back up your stance. If you think it is, you don't understand the EMH.
  - jsprogrammer 11y ago
    >The fact that the information is not real is immaterial to the EMH in any of its forms. Hmm, really? I guess I really don't understand the EMH then. Can you give me a formulation of EMH that doesn't depend on information? I tend to think that the EMH is a tautology and makes no real claims. It's a novelty. Back to our argument though, you're still disregarding the sequence of events. The reputation of the information precedes the information itself. It was possible to know to disregard the content before actually seeing the content. You keep claiming that the content came first, when that is strictly not the case. I will accept that many traders treat any text they come across as gospel and blindly follow it. I'm sure there's a continuum to someone who first verifies all information they depend on in a trade.
    - kasey_junk 11y ago
      I'm not making myself clear. The EMH does require information for any of its formulations. It doesn't require the market to magically determine what is "real" vs "fake" information and disregard "fake" information. In fact quite the opposite. The EMH is based on the idea that lots of people react to incorrect information all the time, but that just as many are working on correct information such that the aggregate is efficient. That is precisely what happened here. Some people reacted incorrectly to new information (ie they believed it) others reacted correctly (ie they saw it was nonsense) and the market in aggregate self corrected. I'm not sure that the correction falls into the semi-strong form as I'm not sure what the boundaries for speed of market reaction are. I think it is smart to be skeptical of the EMH, lots of people are for very good reasons, but this event is not evidence one way or the other.
- prostoalex 11y ago
  A was simultaneous with B only in retrospect. It requires one basic skill (reading) to comprehend A, it requires a variety of skills (understanding of domain name system, technical expertise, knowledge on how to do lookups of a domain owner) to comprehend B. If article A had a large banner saying THIS STORY IS FAKE, you could argue A and B were simultaneous.
  - jsprogrammer 11y ago
    One must have the domain name (and in fact, the entire URL) before one can access the content (A). So, B actually came before A (assuming the article was only accessed over the Web [eg. not also handed out in print on the floor of a prop trade desk]).