Best Execution Compliance Workstation Project Summary
|
Best Execution Compliance Workstation Project Summary
|
|
c Products
Knowledge
Fun
Links
© The Z/Yen Group of Companies 2008
|
The European Union implements the Markets in Financial Instruments Directive (MiFID) on 1 November 2007. Article 21, “Obligation To Execute Orders On Terms Most Favourable To The Client” - ‘best execution’ - of MiFID states:
In the USA, a similar set of regulations from the SEC, RegNMS, takes effect in 2006 and also requires the ability to demonstrate best execution. The Financial Services Authority advises (“Planning for MiFID”, November 2005, Page 12):
Most brokers rely on traditional management oversight of the trading process or customer feed-back to control execution quality – but traditional oversight cannot cope with today’s volumes and clients tend to feed back selectively. Many brokers contrast prices obtained in a sample of trades (1% may be typical according to the British Bankers’ Association) with the published bid-offer spreads available at the time – but then complain that the bid-offer spreads are only a good comparison for very small trades. What’s missing is the ability to show that a specific trade was executed at a reasonable price taking into account the various characteristics of the trade. In order to comply with MiFID, the only effective method of monitoring thousands or hundreds of thousands of trades per week is to have an automated process identifying a sensible set of anomalous trades for individual examination. Basically, firms need a ‘sifting engine’ that puts forward trades that must be examined – “best execution compliance automation”. In 2004 Z/Yen undertook an informal trial of its PropheZy risk/reward prediction software on bid-offer spreads for the small-cap trades of a broker. PropheZy is a commercial application of a Support Vector Machine (SVM). This trial indicated that PropheZy might be good at identifying trading anomalies for compliance purposes. Best Execution Compliance Automation (BECA) Project In 2005 Sun Microsystems and the London Stock Exchange, with the cooperation of four brokers, sponsored a detailed, formal trial of PropheZy using three months of 2004 data comprising over 190,000 trades with a value of over £54bn in order to predict a fourth month. The project objective was to see if PropheZy could predict a number of trade characteristics, in particular the likely price range of a trade (specifically, one of 20 price bands on a logarithmic scale). Other characteristics that were tested for predictability included the counterparty to the trade and the share itself (given all the other characteristics). Chart 1
The project proved that the PropheZy system successfully predicted price movement bands. For instance, by setting the level of acceptable accuracy at “within 0 to 4 bands” out of 20 [Chart 1], i.e. one quarter of the logarithmic scale, PropheZy was able to predict over 50% of the trades price bands acceptably. Using these predictions, it was possible to set a level for best execution outliers or anomalies using price band prediction differences. An outlier, or anomalous trade, was defined as a trade where the predicted price movement differs from the actual price movement by more than 15 bands out of 20 – either a very high price movement was predicted but a low price movement was observed, or a very low price movement was predicted but a high price movement was observed. Table 1 below indicates that when using the SVM as an initial filter [First Filter] on average 7% of non-SETS trades are defined as outliers. 7% is still too many outliers for a detailed manual investigation. A second filter is therefore needed. A trade is unlikely to fail best execution if it was conducted at the best prevailing price (or better), though there are some arguments that very large trades might be capable of exceptional improvement under certain conditions. When excluding trades outside the bid/offer spread [Second Filter] is combined with the first filter the number of outliers that are outside the bid/offer spread is approximately 1%. Table 1 – The number of outliers by broker
The research team also examined how well the SVM could predict the magnitude of a single share’s price movement. By plotting the actual share price movement since the last trade against the SVM predicted price movement band it is evident that the SVM does achieve reasonable predictions of price movement and ultimately responds well to changes in the market. In the chart below the three outliers for the share are plotted in green. It is interesting to note that in all three cases the SVM presages major changes, but for this broker they hadn’t occurred at that point. Chart 2
An inspection of the anomalous trades with the participating brokers was undertaken and it was agreed that the system was providing trades worthy of investigation:
Moving Towards A Compliance Workstation PropheZy is a statistical and information technology approach with applications in numerous areas that Z/Yen terms dynamic anomaly and pattern response. Anomaly detection using PropheZy has wide applicability in a number of trading markets beyond equities, for example foreign exchange, fixed income and commodities. In addition, the sifting approach to identify anomalous trades could be expanded from just price to cost, speed, venue, order fulfilment, client instructions and size. The project was not however, just a statistical exercise. During the course of the project the team provided a prototype “Compliance Workstation”. The Compliance Workstation combined a number of tools (PropheZy, VizZy, FractalIntelligence and Decisionality within an Excel framework) that provided a number of features, specifically the ability to:
Diagram 1: Contrasting a sub-set of actual versus predicted trade price bands
Diagram 2: Drilling down to investigate specific anomalous trades
Diagram 3: Workflow mapping of the investigation process also provides an audit trail
Next Steps The BECA project is exciting because it shows that the automated sifting of trades can identify anomalies, thus reducing the costs of complying with MiFID while simultaneously increasing the effectiveness of the compliance function. The Compliance Workstation is available to firms for purchase in order to help comply with MiFID. The BECA project could be extended, with some further testing, to buy-side (asset managers and investment managers) compliance and other markets. The BECA project also provides an opportunity for exchanges who wish to provide a centralised compliance service for their members. More information:
For further information on this project please contact either Michael Mainelli (michael_mainelli@zyen.com) or Mark Yeandle (mark_yeandle@zyen.com) on +44 20 7562-9562.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
