To solve the problem of poor performance of deep neural network models due to insufficient data, a simple yet effective interpolation-based data augmentation method is proposed: MSMix (Manifold Swap Mixup). This method feeds two different samples to the same deep neural network model, and then...

We examine the non-execution probability and market share of crossing networks (dark pool) using SEC 605 data for NYSE and NASDAQ stocks. We find that crossing networks have lower effective spread and price impact of trade, but they also have lower execution probability and speed of trade....

We present an analysis of user conversations in on-line social media and their evolution over time. We propose a dynamic model that accurately predicts the growth dynamics and structural properties of conversation threads. The model successfully reconciles the differing observations that have...

We examine how fragmentation is affecting market quality in US equity markets. We use newly available trade reporting facilities (TRFs) data to measure fragmentation, and we use a variety of empirical approaches to compare execution quality and efficiency of stocks with more and less fragmented...

type="main" <title type="main">ABSTRACT</title> <p>We investigate odd-lot trades in equity markets. Odd lots are increasingly used in algorithmic and high-frequency trading, but are not reported to the consolidated tape or in databases such as TAQ. In our sample, the median number of odd-lot trades is 24% but in some stocks...</p>

Purpose – The purpose of this paper is to propose a framework for data mining (DM)‐based anti‐money laundering (AML) research. Design/methodology/approach – First, suspicion data are prepared by using DM techniques. Also, DM methods are compared with traditional investigation techniques....