Modeling Trade Direction
Journal of Financial Econometrics, Vol. 10, No. 2, pp. 390-415, 2012
Posted: 8 Oct 2012
Date Written: 2012
I propose a modeling approach to classifying trades as buys or sells. Modeled classifications consider information strengths, microstructure effects, and classification correlations. I also propose estimators for quotes prevailing at trade time. Comparisons using 2800 U.S. stocks show modeled classifications are 1%-2% more accurate than current methods across dates, sectors, and the spread. For Nasdaq and New York Stock Exchange stocks, 1% and 1.3% of improvement comes from using information strengths; 0.9% and 0.7% of improvement comes from estimating quotes. I find evidence past studies used unclean data and indications of short-term price predictability. The method may help detect destabilizing order flow.
Keywords: delay models, flash crash detection, prevailing quotes, trade classification, trade direction, trade publishing delays
JEL Classification: C53, D82, G14
Suggested Citation: Suggested Citation