So far, the scientific literature paints a largely positive picture of this development. Algorithmic trading contributes to market efficiency and liquidity, although the impact on market volatility is still opaque. It is therefore essential to allow algorithmic trading and HFT to develop their advantages in times of serene trading and to have mechanisms (such as circuit breakers) in order to control potential errors both at the level of algorithm users and at the market level. However, preventing the application of these strategies through insufficient regulation, which entails excessive burdens, can have unforeseen negative effects on the efficiency and quality of the market. Trade News (2009). 2000-2009: the decade of e-commerce. Online. www.thetradenews.com/trading-execution/industry-issues/4038 [cited 2012 Jan 17]. Among the first to analyze algorithmic trading patterns in electronic order books, Prix et al. (2007) studied changes in the lifetime of cancelled orders in the XETRA order book. Due to the characteristics of their data set, they are able to identify each task by a unique identifier and thus recreate the entire history of events for each task. Since they focus on the lifetime of so-called wireless delete commands, i.e.
orders inserted without execution and then canceled, they find algorithm-specific characteristics with regard to the insertion limit of an order compared to ordinary human commerce. Gsell and Gomber (2009) also focus on the differences in business models between human and IT merchants. In their data structure, they are able to distinguish between algorithmic and human command transfers. They conclude that automated systems tend to place more but significantly smaller orders. In addition, they show the ability of algorithms to monitor and modify their orders to be at the top of the order book. The authors find that algorithmic trading behavior is fundamentally different from human trading when it comes to the use of order types, the positioning of order boundaries, the behavior of modification or erasure. Algorithmic trading systems use their ability to process data flows at high speed and react immediately to market movements by submitting orders or changing existing orders. Computers monitor market data and possibly other information at very high frequency and return trading instructions based on an integrated algorithm, often in milliseconds“ (p. 1), and Domowitz and Yegerman (2006) find that „[W]e generally define negorithmics as the automated and computerized execution of stock orders through direct market access channels, generally for the purpose of meeting a specific benchmark“ (p. . .