"Forecasting Crashes: Trading Volume, Past Returns and
Conditional Skewness in Stock Prices"
BY: JOSEPH CHEN
Stanford University
Graduate School of Business
HARRISON G. HONG
Stanford University
JEREMY C. STEIN
Massachusetts Institute of Technology (MIT)
Harvard Business School
Document: Available from the SSRN Electronic Paper Collection:
http://papers.ssrn.com/paper.taf?abstract_id=194948
Other Electronic Document Delivery:
http://www.stanford.edu/~chenjs/files/Forecasting_Cras
hes.pdf
SSRN only offers technical support for papers
downloaded from the SSRN Electronic Paper Collection
location. When URLs wrap, you must copy and paste
them into your browser eliminating all spaces.
Date: December 1999
Contact: JOSEPH CHEN
Email: Mailto:chen_joseph at gsb.stanford.edu
Postal: Stanford University
Graduate School of Business
Stanford, CA 94305-5015 USA
Phone: (650) 723-4877
Fax: (650) 725-7462
Co-Auth: HARRISON G. HONG
Email: Mailto:hghong at leland.stanford.edu
Postal: Stanford University
518 Memorial Way
Stanford, CA 94305-5015 USA
Co-Auth: JEREMY C. STEIN
Email: Mailto:jcstein at mit.edu
Postal: Massachusetts Institute of Technology (MIT)
50 Memorial Drive
Cambridge, MA 02142 USA
ABSTRACT:
This paper is an investigation into the determinants of
asymmetries in stock returns. We develop a series of
cross-sectional regression specifications which attempt to
forecast skewness in the daily returns of individual stocks.
Negative skewness is most pronounced in stocks that have
experienced: 1) an increase in trading volume relative to trend
over the prior six months; and 2) positive returns over the
prior thirty-six months. The first finding is consistent with
the model of Hong and Stein (1999), which predicts that negative
asymmetries are more likely to occur when there are large
differences of opinion among investors. The latter finding fits
with a number of theories, most notably Blanchard and Watson's
(1982) rendition of stock-price bubbles. Analogous results also
obtain when we attempt to forecast the skewness of the aggregate
stock market, though our statistical power in this case is
limited.
JEL Classification: G1