Fast and Stable Estimation of the Probability of Informed Trading (PIN)

Utilities for fast and stable estimation of the probability of informed trading (PIN) in the model introduced by Easley et al. (2002) are implemented. Since the basic model developed by Easley et al. (1996) is nested in the former due to equating the intensity of uninformed buys and sells, functions can also be applied to this simpler model structure, if needed. State-of-the-art factorization of the model likelihood function as well as most recent algorithms for generating initial values for optimization routines are implemented. In total, two likelihood factorizations and three methodologies for starting values are included. Furthermore, functions for simulating datasets of daily aggregated buys and sells, calculating confidence intervals for the probability of informed trading and posterior probabilities of trading days' conditions are available.