You are on page 1of 7

3.

10(a)
3.10 (b)
ut =
1 if t is June, July, or August,
0 otherwise,

estimation results show that the estimates of α10 and β10 are statistically
nonsignificant at the 10% level. Therefore, we refine the equation and
obtain the model
rt = 0.0062 + at, at = σt Et ,

σ 2t = 0.00008 + 0.1132a2t−1+ 0.8501σ 2t−1

The Ljung–Box statistics for the standardized residuals


˜at = at/σt show Q(10) = 8.80(1.144) and Q(20) = 17.55(0.6169).
Therefore,
there are no serial correlations in the standardized residuals.
The Ljung–Boxstatistics for ˜a2t
give Q(10) = 6.113(0.806) and Q(20) = 8.229(0.990),
indicating no conditional heteroscedasticity in the standardized residuals
either. The refined model seems adequate.

Because the coefficient 8.004e-5 is significantly different from


zero with a p value of 0.0171, the summer effect on stock volatility is
statistically
significant at the 1% level.

You might also like