Abstract
The term ‘‘price manipulation’’ is used to describe the actions of ‘‘rogue’’ traders who employ carefully designed trading tactics to incur equity prices up or down to make profit. Such activities damage the proper functioning, integrity, and stability of the financial markets. In response to that, the regulators proposed new regulatory guidance to prohibit such activities on the financial markets. However, due to the lack of existing research and the implementation complexity, the application of those regulatory guidance, i.e. MiFID II in EU, is postponed to 2018. The existing studies exploring this issue either focus on empirical analysis of such cases, or propose detection models based on certain assumptions. The effective methods, based on analysing trading behaviour data, are not yet studied. This paper seeks to address that gap, and provides two data analytics based models. The first one, static model, detects manipulative behaviours through identifying abnormal patterns of trading activities. The activities are represented by transformed limit orders, in which the transformation method is proposed for partially reducing the nonstationarity nature of the financial data. The second one is hidden Markov model based dynamic model, which identifies the sequential and contextual changes in trading behaviours. Both models are evaluated using real stock tick data, which demonstrate their effectiveness on identifying a range of price manipulation scenarios, and outperforming the selected benchmarks. Thus, both models are shown to make a substantial contribution to the literature, and to offer a practical and effective approach to the identification of market manipulation.