Research Discussion Papers, Bank of Finland
No 22/2006:
Forecasting market crashes: further international evidence
Terhi Jokipii ()
Abstract: This paper studies the extent to which market crashes are
predictable for a set of six countries, focusing in particular on possible
differences between transition economies (The Czech Republic, Hungary and
Poland) and mature markets (UK, US and EU). We estimate a set of individual
country and pooled specifications to find that market crashes, in the
broader sense, are predictable for all countries analysed. We additionally
investigate the role that investor heterogeneity, proxied by trading
volume, plays in this predictability and find some varying results between
countries. For the Central and Eastern European Countries (CE3), an
increase in trading volume relative to trend appears to have great
predictive power, a result that is supportive of the theory of investor
heterogeneity outlined in the relevant background studies. For the more
mature markets (G5), on the other hand, market crashes appear more likely
to follow a period of increased stock prices and returns, a result fitting
a number of traditional theories, in particular the stochastic bubble
model. Further analysis, allowing for time-varying coefficients, confirms
the volume-crash relationship for the CE3 and provides preliminary evidence
that macro news releases may additionally contribute to the predictability
of market crashes.
Keywords: aggregate market returns; skewness; trading volume; market crash; (follow links to similar papers)
JEL-Codes: C14; G12; G15; (follow links to similar papers)
43 pages, December 14, 2006
Before downloading any of the electronic versions below
you should read our statement on
copyright.
Download GhostScript
for viewing Postscript files and the
Acrobat Reader for viewing and printing pdf files.
Full text versions of the paper:
0622netti.pdf
Download Statistics
Questions (including download problems) about the papers in this series should be directed to Minna Nyman ()
Report other problems with accessing this service to Sune Karlsson ()
or Helena Lundin ().
Programing by
Design by Joachim Ekebom