16
-1
where π≡n0∕n, q is the dimension of both θ1 and θ2, AVar(θ1n0) = (nπ) V1n0, and
л Λ
AVar(θ2,n-no) = (n(1 -π))-lV2,n-no.
We applied this test to the recomposed crystal D1 of the first difference of the 30-
day and the 6o-day interest rates. For that purpose, we fitted a GARCH(1,1) model to the
D1 series for January 2ooo-January 2oo1, and a second model for February 2oo1-April
2oo2. The statistic in (14) was then computed in each case. For both series, we rejected the
null hypothesis of stability of the parameter models. In particular, for both series, we found
that the long-run daily volatility decreased for the second period: from o.77 to o.61 percent
points per day for the 3o-day interest rates, and from o.33 to o.25 percent points per day for
the 6o-day interest rates. As documented by Morandé (2oo2), one effect of nominalization
was to reduce the volatility of nominal interest rates vis-à-vis inflation-indexed interest
rates.
IV Conclusions
In this article, we tested for the presence of structural breaks in volatility by two
approaches: the Iterative Cumulative Sum of Squares (ICSS) algorithm and wavelet
analysis. Specifically, we looked at the effect of the outbreak of the Asian crisis and the
terrorist attacks of September 11, 2oo1 on Emerging Asia, Europe, Latin America and
North America’s stock markets volatility. In addition, we focused on the behavior of
interest rates in Chile after the Central Bank changed its monetary policy interest rate from
an inflation-indexed to a nominal target in August 2oo1.
Our estimation results show that the number of shifts detected by the two methods is
substantially reduced by filtering out the data for both conditional heteroskedasticity and
serial correlation. In particular, for the filtered stock data, the ICSS algorithm did not find