A few months back, Jorge posted a link to a research paper showing a correlation between the mood of tweets and the Dow Jones Industrial Average. The study proved to be interesting, since it aggregated a large amount of random tweets which had nothing to do with the stock market, but displayed a significant correlation.
Twitter mood predicts the stock market
These guys are taking it a step further.
The Derwent Absolute Return Fund, aka The Twitter Fund, seeks to use a similar strategy, following tweets to determine mood and sentiment following a news event. The technology, dubbed The Fourth Dimension, has managed to pull in 25 Million Pounds from investors for the fund, according to CNBC.
Would you invest in the twitter fund? Do you see any potential here?
http://www.cnbc.com/id/%2040793207
You gotta love how the same hedge fund fuck ups who have thumped the Bible of algorithms, phd level stats formulas, multivariate, linear this, brownian that are now creating an investment vehicle purely based on human emotion.
If anything, this is another physical manifestation of mean reversion in the context of not just finance and global economics, but life itself.
This shit should be called the twit fund or the tweety boiyd fucking fund for lisping house cats.
Let's buy/sell purely on the instantaneous reaction of any number of jackasses who may/may not have any rational, logical, qualitative, quantitative, etc...explanation for their maneuvers.
For those who were not born yesterday, this ain't nothing but another attempt to revive the vampires of early finance chat rooms where the last great pump-and-dump scams of the 20th century were carried.
There will be plenty of con men getting rich on this and plenty of suckers swallowing the sword.
But then again... that's the game, is it not?
Good idea. I am going to come up with the "tweeting fund" that posts erroneous tweets and then trades against them.
i wish i had an SB. :(
Haha, I was thinking I would just pay some freshman to pump out a few thousand tweets to effect this rather inane algorithm.
I'm going to start a fund that invests based on the texts submitted to texts from last night. It will apply a formula to each text that determines how much alcohol needed to be consumed to write or experience XYZ event and trade based on the effect this shows for the demand on alcoholic beverages...
haha you could prob branch out to likealittle and gauge condom demand as well
There are some nice possibilities here, add in campuscreepers for pepper spray and we'll have a vice fund in no time.
I've always thought access to social network data could give a trading advantage. For example, I'm sure there is a ton of data in facebook that could be used to extrapolate the popularity of certain products.
Now, if only we could get google and their massive big brother internet watching to start a fund, that would be a killer.
I do think trading based purely on twitter moods might be a little far fetched though
George Soros teaches to trade based on human emotion. He believes that humans are imperfect and often are not trusted to make rational choices. We are constantly trading based on how the market will react to news and market performance is often based on consumer confidence.
Twitter is just another manner in which traders can keep a close pulse on the average american's emotion. Lets face it, the majority of american's do not understand the realities of market participation and therefore make irrational decsions. To understand the mood, confidence and decisions of the american public, becomes an intrinsic tool to an experienced or average trader in the market.
I would rather trade on daily cloud patterns or seasonal elephant migrations.
You can make good money if you're adept at cloud watching!
I dont think this is a bad idea...its just another sentiment gauge of which there are already many (investor surveys, etc) but these type of things can be useful. I have also heard of some analysts incorporating google trends data into forecasts of some economic numbers...for eg alot of people were watching the google search data for "unemployment benefits" because it was very highly correlated to jobless claims. That correlation broke down a bit because of the legislation effecting benefits which caused searches that were not related to actually collecting benefits...but the point is that this type of data can be legitimately useful.
Unde omnis quo neque expedita. Eaque similique velit excepturi nisi. Velit fuga ut et rerum similique cupiditate. Quis et reiciendis voluptas in hic.
Sed reiciendis vel inventore officiis. Excepturi quia esse aliquid porro rerum dolor. Dolorum itaque error numquam eum necessitatibus. Iusto et rerum quidem. Nobis quidem consequuntur sit sequi nesciunt eos accusamus.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...