Empirical results
The share futures onset procedure in the ADEX took place in seven phases, as
illustrated in Table AI, of the Appendix. In Table AII we present descriptive statistics
for the seven onset phases for both the derivative and the control group. Table AII as
well as the corresponding Figure 1, of the Appendix, are useful in order to derive some
ad hoc insights about the first and second moments of the returns of the shares as well
as for the capitalization and trading volume of the corresponding firms. So, the control
sample consists of firms the returns of which have higher standard deviations than
those of the derivative group, while their capitalization and trading volume are lower,
in all derivative onset phases.
InTable AIII of the Appendix we present the estimated logit regressions for the seven
phases of the ADEX. It is obvious that the explanatory power is dominated by the
capitalization and the trading volume variables with a little contribution from
Figure 2.
Time varying coefficients
for the third phase
Time Varying Coefficients - Phase 3
–1.5
–1
–0.5
0
0.5
1
1.5
7/1/2003
7/3/2003
7/5/2003
7/7/2003
7/9/2003
7/11/2003
7/1/2004
Date
Coefficients
Constant
Capitalization
Trading_Volume
Volatility
CGCS
DP
Time Varying p-values for the Capitalization - Phase 3
0.0
0.1
0.2
0.3
0.4
0.5
7/1/2003
7/2/2003
7/3/2003
7/4/2003
7/5/2003
7/6/2003
7/7/2003
7/8/2003
7/9/2003
7/10/2003
7/11/2003
7/12/2003
7/1/2004
7/2/2004
Dates
p-values
Capitalization
Time Varying p-values for the Trading Volume - Phase 3
0.0
0.1
0.2
7/1/2003
7/3/2003
7/5/2003
7/7/2003
7/9/2003
7/11/2003
7/1/2004
Dates
p-values
Trading_Volume
Time Varying p-values for the Volatility - Phase 3
0.0
0.1
0.2
0.3
7/1/2003
7/2/2003
7/3/2003
7/4/2003
7/5/2003
7/6/2003
7/7/2003
7/8/2003
7/9/2003
7/10/2003
7/11/2003
7/12/2003
7/1/2004
7/2/2004
Dates
p-values
Volatility
Time Varying p-values for the CGPI - Phase 3
0.0
0.1
0.2
0.3
7/1/2003
0.4
7/2/2003
7/3/2003
7/4/2003
7/5/2003
7/6/2003
7/7/2003
7/8/2003
7/9/2003
7/10/2003
7/11/2003
7/12/2003
7/1/2004
7/2/2004
Dates
p-values
CGPI
Time Varying p-values for the DP - Phase 3
0.0
0.1
0.2
0.3
0.4
7/1/2003
7/2/2003
7/3/2003
7/4/2003
7/5/2003
7/6/2003
7/7/2003
7/8/2003
7/9/2003
7/10/2003
7/11/2003
7/12/2003
7/1/2004
7/2/2004
Dates
p-values
DP
JFRC
20,3
312
the volatility, CGCS and DP variables. The volatility contributes negatively to the
classification of the two groups. Hence, according to this empirical investigation,
the ADEX board has chosen firms with low volatilities in the derivatives listing.
According to the sign of the CGCS, we conclude that shareholder-friendly (democracies)
firms are better candidates for derivatives listing, than manager-friendly (dictatorships)
firms in some cases. Furthermore, the default probability has a marginal effect in the
derivatives listing process, a fact that is explained by the prevalence of the short run
objectives that the ADEX has been interested in. In order to strengthen the reliability of
the results we have contacted, for each phase of the derivatives listing process, an
analysis of the individual and the joint contribution of the independent variables, as
shown in the corresponding tables of the Appendix.
Finally, we propose an alternative methodology according to which we execute
daily logit regressions for every phase using a rolling window approach. From these
results we conclude with daily estimates of the logit regression coefficients and their
corresponding observed levels of statistical significance. In Figures 2-6 we show the