I have estimated the basic Probability of Informed Trading (PIN) from high-frequency (TAQ) data following Easley, Hvidkjaer, and O'Hara (2002). I apply Lee and Ready (1991) method to infer trade direction by applying a 1-second rule (Henker and Wang, 2006). The PIN measure applied in the robustness section of ourpaper “Dividend Hibernation and Future Earnings: When No Dividend News Is Good News”.
The sample covers all NYSE/Amex common stocks in 2002-2014 for which I could obtain estimates.
The data sample is available at request: please send an email to Bardia Khorsand.
I construct the indexes from several adverse selection and illiquidity measures developed by the microstructure strictures from low-frequency data, available at quarterly and yearly frequencies.
At June each year, I form portfolio of Zombie firms, defined as non-financial US firms with high book-to-market ratios that appear for more than 10 years in the CRSP tapes and report the interest coverage ratio below one for past three consecutive years. Then, above (below) median NYSE market Cap firms are allocated to Small and Big, and firms at top (bottom) 30% of the Book-to-Market deciles are assigned to High (Low). The middle 40% percent represents Medium (Neutral). I hold zombie firms in the portfolio until July next year when the portfolios are reformed.