Tossicità del flusso degli ordini e microstrutture del libro degli ordini di cryptovalute

The main goal of this thesis the is to dive deep into how the order book for Bitcoin actually works. We want to understand how the market behaves under pressure and gain valuable insights into its dynamics. To achieve this, we'll be using a measure called the VPIN indicator , which will help us to figure out how toxic the order flow is. This indicator will give us a glimpse into how market participants react and adapt when there is an imbalance between the volume of buy and sell orders. We'll also explore different models to see which ones can accurately capture the complexities of the Bitcoin market. These models will take into account factors like liquidity, price impact, and market depth. By comparing these models, we hope to improve our understanding of how the Bitcoin market really works. In addition to analyzing the order book and its statistical properties , we'll be studying real trades to get a complete picture of what's happening in the market. We'll use a self-exiting process called Hawkes process to model the trade data, which allows us to observe patterns and clusters of activity over time. This analysis will give us valuable insights into how trades are connected and influence each other. We will use also a neural network to try to predict, based on the indicators provided by the analysis above, the return of the next bucket. Overall, the aim of this research is to provide valuable insights into the inner workings of the Bitcoin market. By studying its microstructure and carefully analyzing trade data, we hope to contribute to the existing knowledge about cryptocurrency markets and deepen our understanding of how they behave