In Finance

This dissertation consists of two essays. The first essay develops a new methodology for estimating the probability of informed trading from the observed quotes and depths, by extending the Copeland and Galai (1983) model. This measure (PROBINF) can be computed for each quote and it represents the specialist’s ex-ante estimate of the probability of informed trading. I show that PROBINF exhibits a strong and robust relationship with the observed level of insider trading and with measures of the price impact of trades (λ) estimated based on the models of Glosten and Harris (1988), Madhavan and Smidt (1991) and Foster and Viswanathan (1993). In contrast, the alternative measure of the probability of informed trading (PIN) developed by Easley, Kiefer, O’Hara and Paperman (1996) exhibits a weaker and less robust relationship with insider trading and price impact of trades. The time series pattern of PROBINF in an intra-day analysis around earnings announcement is consistent with previous findings regarding informed trading. An important advantage of PROBINF over PIN and other measures of information asymmetry such as price impact of trades and adverse selection component of the spread is that, unlike these measures, it can b