Tweets that beat the market

  • Published on

  • View

  • Download


24 | NewScientist | 10 September 2011TECHNOLOGYSocial-media opinions about cars are particularly good at predicting Ford and GM stock pricesEVER been clothes-shopping but struggled to find the perfect fit? Measuring garment designs against thousands of laser-scanned body shapes could soon solve that problem by helping designers tailor their sizes to fit the majority of people. Average size charts are traditionally created by measuring the distances between key points on a number of peoples bodies and tabulating the measurements. Artists then create a computer model to fit these sizes. Based on a few numbers you cant get an accurate 3D shape that you can use, says Stefanie Wuhrer a t Saarland University in Saarbrcken, Germany.She and colleagues have a system that lets clothes manufacturers determine how many sizes of an item they will need to stock to fit as many members of a population as possible.Designers start with the relevant body measurements and the extent to which they can vary based on a database of over 5000 3D scans that describe the range of body shapes in a chosen population. For example, the three key measurements for a helmet, say, head width, depth and height, can vary by up to 3 centimetres.Using all the relevant data, the system calculates that a designer would only need three helmet sizes to cover 95.2 per cent of people in a target population. This suggests a manufacturer could please most of their customers with just a few designs.Nick Clarke, who researches computer-aided design for fashion retailing at the University of Manchester, UK, says the model is an interesting idea but quite theoretical and would require testing before manufacturers could use it. Jacob Aron nTHE trend is your friend, as they say on Wall Street. But when it comes to financial decisions, can you trust a Twitter trend? Possibly: an analysis of sentiments expressed on Twitter appears to have given the small London-based firm Derwent Capital Markets an edge. Derwents 25 million fund finished its first month of trading in July with a return of 1.85 per cent. By contrast, the Standard & Poors 500 financial index fell 2.2 per cent and the average hedge fund made only 0.76 per cent.Beginners luck? Perhaps. Or maybe the tweets are helping. Derwents system tracks emotions expressed across 10 per cent of the roughly 100 million daily tweets using algorithms devised by Johan Bollen, a computer scientist at Indiana University Bloomington. It then uses this information to predict changes in the stock market.In a study published last year, Bollens algorithms predicted the direction of the daily swing of the Dow Jones closing price with 87.6 per cent accuracy. The index consistently rose a few days after a period of calm tweets and dipped a few days after a period of anxious tweets ( Derwent is not alone in its approach. Sentiment analysis makes a huge amount of sense to a lot of people there are a lot of companies looking into this, including Reuters, Dow Jones and many start-ups, says Seth Grimes, an analyst who chairs the Sentiment Analysis Symposium.Recently, financial firm Bloomberg partnered with WiseWindow, based in Irvine, California, which claims to offer a near direct correlation between social media conversations and stock price by aggregating information from Twitter, Facebook, blogs and such. Unlike Derwent, WiseWindow sorts the opinions it collects online by industry. In one test, WiseWindow found that opinions about cars are particularly good at predicting the stock prices of Ford and General Motors, for example.Will sentiment analysis of tweets and blogs revolutionise stock market forecasting? No. Marketers and investors have been surveying the publics mood for ages through focus groups and surveys. The only thing that has changed is that social media and powerful analytical software now provide a way to keep constant tabs on how the masses are feeling. Improvement in our predictions of stock market movements will be small. But even a 1 or 2 per cent increase in profit represents huge sums for multibillion-dollar investment firms.What were doing now could only have happened because of social media, says Sid Mohasseb, CEO of WiseWindow. The fact that it works shouldnt be so astonishing to us its still just people. People are saying, I want this, I dont want that and their choices drive revenue. We are just processing it much faster. Ferris Jabr nDigital tailor gives the perfect fit, every timeInsIghT Financial futuresHow to tweet the marketUsing social media to predict the stock market is all the rage. Will it last?