The TRADE 63 - Q1 2020 | Page 11

[ T H O U G H T L E A D E R S H I P | G O L D M A N S A C H S ] DECIPHERING GLOBAL EXECUTION DYNAMICS FOR OPTIMAL TRADING I n today’s increasingly com- plex and inter-related market environment, (i) information dissemination is occurring almost instantaneously around the world, (ii) market participants and financial products are more global in nature, creating or reshaping linkages across markets that didn’t exist just a few years ago and (iii) the electronifica- tion of trading is prevalent across most asset classes. Against this backdrop, algorithmic approaches to execution are at an arms race of consuming enormous amounts of data, coupled with the need to deploy more sophisticated models to process, stratify and produce useful forecasts out of such wealth (or chaos) of information. The quintessential pillars of any execution modelling framework are risk and transaction cost. More passive, less impactful execution re- quires trades to be more spread out, thus carrying higher time risk. Con- versely, to execute rather quickly one would incur higher cost/induce higher impact. The question that naturally arises, is, “how can we better model risk and cost to optimally decide how much and how fast one should trade”? This quest of continuously improving risk and cost models, as well as optimisation processes, lies at the heart of algorithmic research. Below we highlight some of the topics that our team, the Quanti- tative Execution Services arm of Goldman Sachs, has been tackling and share some of our findings and observations. Several market micro and macro challenges have changed the way we think about and model execu- tion dynamics, be they regulatory, competitive pressure or economi- cally driven. A tremendous amount of our research effort is directed towards solving the micro challeng- es including placement logic, venue selection, latency, high frequency signal research etc. In this article we deal with the more ‘macro’ effects— of minute-frequency and lower. Some of the key themes that have emerged over the last decade are: The shift from active to passive investing – the rise of ETFs and other index tracking investment vehicles to the detriment of active investment management is a well-documented phenomenon. Our research suggests that there is a statistically significant relation- Michael Steliaros, Global Head of Quantitative Execution Services, Goldman Sachs ship between passive fund and ETF ownership and volatility, specifically towards the end of the trading day, the closing auctions themselves and subsequent overnight dynamics. This is not necessarily apparent when looking at simple, close-to- // thetradenews.com Issue Issue 63 63 // thetradenews.com // // 11 11