The name is absent

Hansen and Johansen (1993) provide a method to analyze not only the extent but also the
dynamics of the long run relationships. Their recursive cointegration approach relies on Johansen
and Juselius (1990) cointegration test. Recursive analysis is performed for an initial period and
thereafter updates as new data are added are added to the initial sample. Consequently, the
statistic of interest is calculated over the chosen sample, say t0 to tn. This sample is then extended
by j periods and the statistic is re-estimated for the period from t0 to tn+j.. Eventually, the
estimation procedure reaches the end of the data, producing the test statistic results equivalent to
the standard static Johansen and Juselius (1990) estimation over the entire time period. The
relevant trace statistic is then plotted and examined for interpretation. For ease of interpretation,
the calculated trace statistic is rescaled to a critical value, usually 90 % or 95 %. Rescaled values
above 1 of the trace statistic for the null hypothesis of τcointegration relationships against k
cointegration relationships indicate against the null hypothesis. In case of the null hypothesis of
no cointegration relationships, an upward trend indicates either increased integration and/or a
move toward integration; a downward trend indicates decreased integration and/or a move away
from integration.

5.3. Dynamic Conditional Correlations GARCH (DCC-GARCH)

We also use the recent dynamic conditional correlation specification of multivariate
GARCH models (DCC-GARCH; Engle, 2002) to model our time series. Analysis of correlations
between international asset markets has been a cornerstone for making inferences about the short-
term interdependencies between the markets and presence of diversification benefits (Grubel
(1968), Longin and Solnik (1995)). Earlier studies relied on analysis of simple correlation
coefficients (see for example Panton, Lessig and Joy (1976) and Watson (1980) whereas later
studies utilized rolling correlation coefficients and correlation coefficients adjusted for the

15

<<   14   15   16   17   18   19   20   >>

More intriguing information