TY - JOUR
T1 - Japanese stock movements from 1991 to 2005
T2 - Evidence from high- and low-frequency data
AU - Nagayasu, Jun
N1 - Funding Information:
The author would like to thank Michael Clements, Jiti Gao, Lars-Erik Oller and Isabel Casas Villalba for the useful comments and discussion. An earlier version of this article, ‘Predicting high frequency Japanese stock market data,’ was presented at the 24th International Symposium on Forecasting in Sydney, Australia (2004). I also appreciate the comments from the editor and anonymous referees that helped improve this article. All remaining errors and omissions, however, are my responsibility. This research was initiated when the author was an IMF staff member and was supported by Ishii Memorial Foundation for Equity Research (Ishii Kinen Shoken Kenkyu Shinko Zaidan).
PY - 2008/3
Y1 - 2008/3
N2 - This article analyses movements in Japanese stock returns in the recent period (1991-2005). Unlike previous literature, by modelling persistence in both the mean and volatility simultaneously, first, we find evidence of persistence in Japanese stock returns. Second, while not so for daily returns, changes in monthly returns are found to reflect those in economic fundamentals, such as the interest rate and the dividend-price ratio. This finding is consistent with the conventional belief that higher frequency returns tend to move more in response to non-economic fundamentals, and we confirm that the poor long-term performance of Japanese stocks can be explained by economic factors. The statistical models are thoroughly examined by diagnostic tests in the contexts of the in- and out-of-sample forecasting.
AB - This article analyses movements in Japanese stock returns in the recent period (1991-2005). Unlike previous literature, by modelling persistence in both the mean and volatility simultaneously, first, we find evidence of persistence in Japanese stock returns. Second, while not so for daily returns, changes in monthly returns are found to reflect those in economic fundamentals, such as the interest rate and the dividend-price ratio. This finding is consistent with the conventional belief that higher frequency returns tend to move more in response to non-economic fundamentals, and we confirm that the poor long-term performance of Japanese stocks can be explained by economic factors. The statistical models are thoroughly examined by diagnostic tests in the contexts of the in- and out-of-sample forecasting.
UR - http://www.scopus.com/inward/record.url?scp=40449103015&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=40449103015&partnerID=8YFLogxK
U2 - 10.1080/09603100600675490
DO - 10.1080/09603100600675490
M3 - Article
AN - SCOPUS:40449103015
SN - 0960-3107
VL - 18
SP - 295
EP - 307
JO - Applied Financial Economics
JF - Applied Financial Economics
IS - 4
ER -