A new PIN model with application of the change-point detection method

The article presents a new Probability of Informed Trading (PIN) model that addresses the shortcomings of existing models by employing change-point detection to adaptively determine the number of information types. This approach allows the model to better fit empirical data and account for autocorrelation in order flows, which is not considered in traditional models. The model is validated using Apple stock data and shows significant variance in the estimated number of information types, highlighting its superior performance. Additionally, the model enables the explicit identification of information impacts, providing insights into the time-related aspects of information transfer within the market. The article concludes by emphasizing the potential of this new PIN model for future research and applications in the field of financial econometrics.