Stock market manipulation is detrimental to traders and corporations, causes unnecessary price fluctuations, and only benefits financial criminals. The research presented here determines an appropriate model to help identify stocks witnessing activities that are indicative of potential manipulation through three separate but related studies.

In Developing an Effective Model for Detecting Trade-Based Market Manipulation, classifiers based on three different techniques namely discriminant analysis, a composite classifier based on Artificial Neural Network and Genetic Algorithm and support Vector Machines is proposed. The proposed models help investigators, with varying degree of accuracy, to arrive at a shortlist of securities which could be subject to further detailed investigation to detect the type and nature of the manipulation, if any.

Following a fluid outline, Developing an Effective Model for Detecting Trade-Based Market Manipulation, introduces the topic, explores the aims and scopes of the research, before delving into the data and modelling to explore their application to the stock market to detect price manipulation.