25.06.2013 Views

statistique, théorie et gestion de portefeuille - Docs at ISFA

statistique, théorie et gestion de portefeuille - Docs at ISFA

statistique, théorie et gestion de portefeuille - Docs at ISFA

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

1.4<br />

1.2<br />

1<br />

0.8<br />

0.6<br />

0.4<br />

0.2<br />

0 1 2 3 4 5 6 7<br />

x 10 −3<br />

0<br />

log−r<strong>et</strong>urn x<br />

10 1<br />

10 0<br />

10 −1<br />

10 −2<br />

10 −5<br />

10 1<br />

10 0<br />

10 −1<br />

10 −3<br />

y ∼ x 0.54<br />

10 −4<br />

10 −2<br />

log−r<strong>et</strong>urn x<br />

y ∼ x 0.77<br />

Figure 12: Upper panel: Sample tail (continuous line), local in<strong>de</strong>x β(x) (dashed line) and local exponent c(x)<br />

(dash-dotted line) for the neg<strong>at</strong>ive tail of the Nasdaq five minutes r<strong>et</strong>urns. Lower panel: doubled logarithmic<br />

plot of the local in<strong>de</strong>x β(x). Over most of the d<strong>at</strong>a s<strong>et</strong>, β(x) increases with the log-r<strong>et</strong>urn x as a power law<br />

of in<strong>de</strong>x 0.77 while beyond the quantile 99% (see the ins<strong>et</strong>) it behaves like another power law of a smaller<br />

in<strong>de</strong>x equal to 0.54. The goodness of fit with these two power laws has been checked by a χ 2 test.<br />

69<br />

10 −3<br />

10 −2<br />

133

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!