Computing Science: Volatility Management of High Frequency Trading Environments.

摘要

High frequency trading (HFT) environments provide technologies that enable algorithmic trading within automated marketplaces. The most prominent example of an HFT environment is within equity trading, where many millions of trades are achieved at a high volume to gain a reasonable cumulative profit. Such environments rely on low latency/high performance technologies to allow trades to react in a timely manner to market volatility. However, sometimes the volatility of the market goes beyond what supporting infrastructure can allow, resulting in erroneous behaviour of the marketplace. In this paper we tackle the problem of managing market volatility to limit erroneous market behaviour. Our approach is unique in that it is non-dependent on the trading environment itself and self-regulates based only on trading frequency and contention. We demonstrate our results and show that by managing trade injection rates and contention of shared state the volatility of HFT environments can be managed appropriately and in an automated manner.