|
Accession number;99A0263652
|
| Title;Observation of Rational Bubbles by Independent Component Analysis. |
| Author;
MAEKAWA SATOSHI
(Commun. Rese. Lab.)
FUJIWARA YOSHIHISA
(Commun. Rese. Lab.)
|
Journal Title;IEIC Technical Report (Institute of Electronics, Information and Communication Engineers)
|
Journal Code:S0532B
|
ISSN:0913-5685
|
|
VOL.98;NO.499(AI98 68-75);PAGE.17-22(1999)
|
| Figure&Table&Reference;FIG.7, REF.8 |
| Pub. Country;Japan |
| Language;Japanese |
| Abstract;We observe from the joint probability distribution for successive logarithmic returns of daily stock prices that they are statistically dependent. We apply independent component analysis to decompose the stock price changes into independent components, and obtained the impulse response for each component. Because the impulse responses are almost orthogonal, we model it with a rotation filter. The orthogonality is compatible with vanishing autocorrelation function. On the other hand, from the phase information, we show that the rotational model can explain a long-term fine structure of the impulse response for decades of days. Thus, although the daily returns are not independent, long-term efficiency in market holds. We believe that this work observes overshoot of price change, or the so-called rational bubble. This observation might contain a useful information on heterogeneity of information and presence of arbitragers in an efficient market. (author abst.) |
|
|
|
Related Articles;
|
