文档介绍：Market Microstructure and
Stock Return Predictions
Roger D. Huang
Hans R. Stall
Vanderbilt University
To what extent are the empirical regularities
implied by market microstructure theories useful
in predicting the short-run behavior of stock
returns? A two-equation econometric model of
quote revisions and transaction returns is devel-
oped and used to identify the relative importance
of different microstructure theories an& to make
predictions. Microstructure variables and fagged
stock index futures returns have in-sample and
out-of-sample predictive power based on data
observed at five-minute intervals. The most strik-
ing microstructure implication of the model, con-
firmed by the empirical results, specifies that the
expected quote return is positively related to the
deviation between the transaction price and the
quote midpoint while the expected transaction
return is negatively related to the same variable.
The burgeoning literature on the microstructure of
securities markets contains substantial evidence of
systematic behavior in the short-run pattern of stock
prices. Prices exhibit reversals as they bounce between
We have benefitted from ments of Bill Christie, David Easley, Bob
Hodrick, Bob Korajczyk, Ron Maculis, Mark Ready, Peter Rossi, e Soli-
anos, Erik Sirri, Chester Spatt (executive editor), and seminar participants
at University of Alabama. Northwestern University, Notre Dame University,
University of Utah Winter Conference, USC/UCLA/NYSE Conference on
Market Microstructure, Dartmouth College, Harvard University, European
Finance Association Conference, and Vanderbilt University. We are especially
grateful for the ments of the referee, Ananth Madhavan, and
the editor, Andrew Lo. This research was supported by the Dean’s Fund for
Research and by the Financial Markets Research Center at the Owen Grad-
uate School of Management, Vanderbilt University. Address correspondence
to Roger D. Huang.