Анализ PIN меры с помощью теоретической модели

Informed trading is not directly observable but has significant implications for the asset prices. Different metrics are intended to measure informed trading, but the most widely used is Probability of Informed Trading (PIN). There is a long-standing discussion about whether PIN reflects informed trading or not. In this paper I test the relation between PIN and the proportion of informed trading and conclude that PIN does not measure informed trading. I use a theoretical model of informed trading which allows controlling for economic variables: prices, volumes, whether a trade occurred due to information or liquidity reasons. The literature demonstrates that this theoretical model of trading shares features of empirical data. Simulations of the model are used to generate artificial data and test the PIN with it. Also, I propose and prove statements which show properties of the asset value variance condition on the market makers' information.