|
Every trader knows that not all trades are "created equal". In some cases trader knowledge and expertise will have little impact on the execution outcome. At the other extreme are trades where the trader and the execution strategy they follow can and do make an enormous difference. Traders use their experience to help identify trades in the latter group while trying to minimise the amount of time spent on the former. Landscape Research and Analysis is the latest innovation from GSCS that allows traders to create a process for identifying where the opportunities for improvement lie. It uses sophisticated custom analysis of post-trade data to create a pre-trade strategy recommendation within the trader's order management system. Landscape also allows traders to easily explain to portfolio managers and others, the framework within which they have to operate. Landscape is a forward chaining expert process that is completely customised to each individual client. It consists of three stages each of which leverages the unique database and reporting capabilities of the itero system. The first stage is to construct a series of Trading Landscapes that accurately describe, and pictorially represent the trading environment. These Landscapes are based around the important factors that typically determine the difficulty of any particular trade execution; the size of the transaction, the momentum of the market and security, the volatility and the bid/ask spread. Chart One illustrates such a Landscape which focuses on momentum and volume characteristics taken from the itero database. GSCS works with the client to assess which of the various Trading Landscapes are most relevant for them. For example a small manager might not regard trade size as being relevant to their trading assessment, but may be more interested in volatility. Chart One  click to view a larger version Once the appropriate Trading Landscape(s) are identified, GSCS will present the associated Performance Landscapes. This makes use of the multiple benchmark analysis available in itero. The Performance Landscapes (two examples of which are shown in Chart Two that highlight performance against Release Time Price and Available VWAP) show the client where in the Landscape, real difference in results is available. These Landscapes are normally viewed based on absolute trading costs and gains (as shown here) but can be presented as percentages as well. Depending on the number of Trading Landscapes and relevant benchmarks, there could be perhaps thirty different Performance Landscapes that GSCS would take into consideration. Chart Two  click chart for larger version
Chart Three  click chart for larger version The final part of the Research involves GSCS working with the client to decide how transactions should be categorised within the OMS and what data will be needed in real-time to correctly assign transactions between these categories. For each category a "recommended" trading strategy will be highlighted. Depending on the level of sophistication of the OMS and the desires of the traders and the client, this recommendation may simply be recorded for future analysis or could be made difficult to override. Strategies could include anything from DMA through a volume participation algorithm or worked order. The latter would identify those trades from which the firm would derive greatest potential benefit from trader time. The overall Landscape process involves original thought and analysis by GSCS using itero data for the client and, where appropriate, comparing it to the itero Universe of data. The information is not generally available, the analytical tools are proprietary and the research is client specific not generic. The outcome is designed specifically to enhance the trading performance and improve both investment returns and the efficiency of the trading desk. The design of the system by GSCS deliberately allows for the possibility that its use could be extended to evaluation of broker performance. However this will be determined by the level of client interest. |
 |
|
Landscape
