I tweeted an article yesterday about a study to predict fluctuations in the stock market by analyzing twitter moods about trending topics related to corporate performance.
-----------------------------------------------------------------------------------------
"So these guys took 9.7 million tweets posted by 2.7 million tweeters between March and December 2008 and looked for correlations between the GPOMS indices and whether Dow Jones Industrial Average rose of fell each day."
"We find an accuracy of 87.6% in predicting the daily up and down changes in the closing values of the Dow Jones Industrial Average," say Bollen and co"
-----------------------------------------------------------------------------------------
It makes sense. After all, a portion of the Twitter community is not Beiber obsessed and probably maintains opinion and invests in any number of industries.
Playing the stock-market sucks. Back when I was designing toilets as an intern at a small RV manufacturer in my home-town, I dappled in day-trading, shorting and making 1-2 month long investments in all sorts of stock. My interests were in Technology, however my biggest success was with Pharmaceuticals and Mining.
I used to listen to earnings calls and hedge on overnight results via shorting or bidding, I was rarely met with success. For some reason, I never got it right. I'd say "up" and everyone else (with considerably more market pull then me) would say "down".
What gave? Once I hedged on a pharmaceutical company via an anonymous tip on the Bull-Board that the FDA had approved one of their ingredients in a topical foot-cream. The next day was spent away from my computer, testing toilets for fitment. On that day, the stock value inflated to over 700% of the previous day's closing price. By the time I was at my computer, the bomb was dropping. Fortunately I managed to get out with a considerable gain of just over 410%.
6 months of analysis and education were validated in 8 hours. I had paid my "trading tuition" and it all came down to luck. This was the last time I bought stock.
Where are you going with this?
4 Years have passed and last night I found myself laying in bed, listening to Steve Jobs' rant about tablet screen sizes and sandpaper during Apples Q4, 2010 earnings call.
The big announcement? $20B dollars. My instinct? BUY!!! Reality? See below, October 19th, 2010.
Can Social Media be leveraged to predict or quantify stock prices in advance? Perhaps this is the software-messiah I've been waiting for.
I've also heard that Social Media has been used to predict Box-Office sales.
Is a solid purpose emerging for these networked technologies? Can advanced algorithms identify success, failure and fluctuations in stock prices? What about global economy?
This is very exciting stuff, especially for you Data nerds. Where are the apps? Where is my data visualization?
