Does more data mean better judgement?
Jinfo Blog

9th August 2007

By Tim Buckley Owen

Item

According to the latest Global Market Data Report from IRN, the industry is set to grow by 40% by 2011. Reference data – including news, comment and research on companies – will be the main growth segment, and algorithmic trading will be one of the factors driving it. (More details at http://www.irn-research.com/index.php?/main/news/market_data_report). We looked at algorithmic trading last month – http://web.vivavip.com/forum/LiveWire/read.php?i=36 – and considered what it might mean for flesh and blood information professionals. ‘Every scrap of information and data on a market, commodity, company or CEO is being used to make judgements about buying and selling securities and commodities, investment banking and wealth management,’ IRN reports. ‘Such data allows more judgement factors to be built into computer trading systems – information rather than data is now driving this market,’ adds IRN Director Gary Giddings. But is he right? A recent Economist Buttonwood feature may suggest otherwise (http://www.economist.co.uk/finance/displaystory.cfm?story_id=9482952). It refers to a study that gave horse-racing handicappers varying amounts of information when ranking horses. Not surprisingly, the more information they received, the more confident they became about their answers. But the success of their predictions was actually worse when given 40 pieces of information than when given five. As Buttonwood reports, ‘More information does not necessarily lead to better decisions’. Analysts tend to go with the herd, Buttonwood continues. ‘They gorge on a daily diet of supposedly important economic data. Most of the statistics are revised in subsequent weeks but the revisions rarely have as much market impact as the original figures.’ These comments – and others by Mike Foster writing in Wealth Bulletin (http://www.wealth-bulletin.com/article_detail.php/1123/the_dangers_of_making_predictions) – are prompted by a recent book by fund manager Richard Oldfield. In Simple But Not Easy, he contends that ‘extrapolation is one the fatal flaws of stock market behaviour’. ‘Price movements in response to news are exaggerated, providing an opportunity to those who do not base too much on what has happened in the last hour or 24 hours,’ Oldfield suggests. It could be bad news for both analysts and algorithmic trading if he’s right – but at least one blogger disputes this view. ‘This statement is in contradiction to an abundance of research done on Post Earnings Announcement Drift (PEAD): the phenomenon that stocks with positive earnings news continue to trend up for a long time,’ states quantitative trading consultant Ernest Chan in his blog – http://epchan.blogspot.com/2007/07/more-on-news-driven-trading.html. In any case, the book that caused all the flutter seems to be rather hard to come by. You could keep an eye on http://digbig.com/4tjjn. But despite its recent publication date, Amazon reported on August 9 that it was currently unavailable. ‘We don't know when or if this item will be back in stock,’ Amazon added. Is there scope for some conspiracy theorising here perhaps?

« Blog