Advances in Risk Management

RICCARDO BRAMANTE AND GIAMPAOLO GABBI 235

150

100

50

0

 50

 100

100 200 300 400 500

Number of observations  558 daily changes

Spread between actual and estimatedRESID

dependent variable

No. of observations

Figure 12.8Plot of residuals of correlation USD–JPY changes explained by

volatility differences

Table 12.7Regression equation of correlation USD–JPY changes explained
by volatility differences

Variable Coefficient Std. error t-statistic Prob.

DJPYV1 252.7362 131.7969 1.917619 0.0605
DJPYV5 276.4569 144.7156 1.910346 0.0614

R-squared 0.099577 Mean dependent var 13.08877

AdjustedR-squared 0.082902 S.D. dependent var 24.74389
S.E. of regression 23.69604 Akaike info criterion 9.203554

Sum squared residuals 30321.12 Schwarz criterion 9.275888
Log likelihood −255.6995 F-statistic 5.971798

Durbin–Watson stat. 2.022590 Prob(F-statistic) 0.017834

Notes: The dependent variable; differences of exponential correlations between the equity US market
and the equity Japanese market; Explanatory variables are DJPYV1: differences of exponential volatility
of the equity Japanese market with 1 day lag; and DJPYV5: differences of exponential volatility of the
equity Japanese market with 5 days lag. Number of observations=56 daily changes.

than the corresponding errors for the whole time series. It is necessary to
note, on the other hand, that the correlation higher changes analysis refers
to series composed of only 10 data points and it is not useful in predicting
the time point in the future where these changes will occur.