Advances in Risk Management

RICCARDO BRAMANTE AND GIAMPAOLO GABBI 233

RESID

510152025303540455055

10

8

4

2

6

0

 2

 4

Number of observations  56 daily changes

Spread between actual and estimated

dependent variable

No. of observations

Figure 12.7Plot of residuals of correlation EUR–JPY higher changes

explained by volatility differences

Table 12.4 Regression equation of higher correlation EUR–JPY changes
explained by volatility differences

Variable Coefficient Std. error t-statistic Prob.

DEURV7 22.98515 5.336414 4.307229 0.0001
DJPYV6 17.61808 5.908526 2.981807 0.0044

R-squared 0.113741 Mean dependent var 1.296226
AdjustedR-squared 0.096698 S.D. dependent var 1.472966
S.E. of regression 1.399940 Akaike info criterion 3.547069

Sum squared residuals 101.9112 Schwarz criterion 3.620735
Log likelihood −93.77087 F-statistic 6.673604

Durbin–Watson stat. 1.920604 Prob(F-statistic) 0.012636

Notes: The dependent variable; differences of exponential correlations between the equity euro mar-
ket and the equity US market. Explanatory variables are DEURV7: differences of exponential volatility
of the equity euro market with 7 days lag; DJPYV6: differences of exponential volatility of the equity
US market with 6 days lag. Number of observations=bigger 56 daily changes.

tequals 0.286315 in Table 12.5). Table 12.6 contains the regression estimates
explained only by differences of exponential volatility of the equity US mar-
ket and Figure 12.8 displays the plot of residuals of correlation between
the USD–JPY changes explained by volatility differences. As in the other
cases, Table 12.7 shows how statistics improve using only the highest corre-
lation changes between the US and the Japanese market. Figure 12.9 displays