![]() |
see beyond the numbers in algo tradingAlgo trading is the use of computer programs for entering trading orders with the computer algorithm deciding on certain aspects of the order such as the timing, price, or even the final quantity of the order. Current algo trading systems are mostly quantitative, i.e. they perceive the markets through numbers in time series. But there is more to the financial markets than p/e ratios and betas. In order to be effective, human traders need to take into account qualitative information about market sentiment, financial, technological, and business innovations, political and economic developments, etc. This is the kind of information you would find out about by reading Reuters, Dow Jones, or Bloomberg news. But your computer can't do that ...yet. This is why we believe that the next generation of algo trading systems should be equipped with natural language information processing (NLP) capabilities, and in fact there have been reports in the economist and the banker, that the likes of Credit Suisse, Morgan Stanley, JP Morgan and Dresdner Kleinwort are already using this kind of technology. This is a highly complex bit of technology. It is not enough for your computer to spot and act on keywords. It would have to understand the meaning of any given news item, i.e. its economic and financial significance and its impact on the market. ...but thanks to recent advances in NLP, and in computational semantics in particular, this is now becoming possible. enforce policy & regulatory complianceFinancial institutions are under strict regulatory requirements, concerning what kind of information can flow between what people within the organization. But the IT systems which provide the infrastructural backbone for information exchange in most organizations are usually ignorant of them. This is because, there is currently no way for, say, your e-mail system to understand the content of a message you might be sending to some colleague. Consequently, your computer doesn't know when you are about to breach a Chinese wall. Again: Thanks to recent advances in NLP, this is now moving into the range of possibilities of what your computer can do for you by understanding and intelligently acting on information represented in natural language. |