13
increases by INR 142.67 per tonne while for every unit
increase in value variable of the coal blocks, the bid amount
increases by INR 859.29 per tonne. Both these estimates
have low p-values and, thus, are statistically significant.
The result of tests for Excess Bid shows that for
every unit increase in number of bidders, the Excess Bid
(bid amount over and above the comparable CIL prices)
increases by INR 164.87 per tonne while for every unit
increase in value variable of the coal blocks, the Excess bid
amount increases by INR 9.46 per tonne. The estimate for
the impact of competitive intensity on Excess Bid has low
p-value and, thus, is statistically significant. But the esti-
mate for the impact of Value variable on Excess Bid has
high p-value and, thus, is not statistically significant, indi-
cating that Value variable does impact Excess Bid in statisti-
cally significant way.
Result of Testing the Moderator Model
We ran the test for the Moderator Model, as in the Base
Model, with two scenarios of dependent variable being the
Bid amount and Excess Bid. This was to test the hypotheses
3 that is to estimate the moderating effect of competitive
intensity on relationship between the dependent variable
with value. In the implementation of the test, apart from
competitive intensity and value variables as independents,
we have an interaction term of both these independent
variables. The estimate of the coefficient of this interaction
term and its statistical significance are of our interest. The
results of the tests are presented in Table 5.
The results indicate that for Bid amount, the competi-
tive intensity does not appear to have a moderating effect
on the relationship between Value and Bid amount, even
though competitive intensity does impact the Bid amount
in statistically significant manner. The results indicate simi-
lar outcome for Excess Bid as well, the competitive inten-
sity does not appear to have a moderating effect on the
relationship between Value and Excess Bid, even though
competitive intensity does impact Excess Bid in statistically
significant manner. It is worthwhile to note that in both
cases of Bid amount and Excess Bid being the dependent
variable, when we test of moderation effect by including an
interaction term, the estimate for Value variable impact-
ing the dependent variables is seen to have no statistical
significance. That provides explanation for the outcome of
test that competitive intensity does not appear to impact
their relationship.
Result of Testing the Mediator Model
We ran the test for the Mediator Model, as in the Base and
Moderator Models, with two scenarios of dependent vari-
able being the Bid amount and Excess Bid. This was to test
the hypotheses 4 that is to estimate the mediating effect of
competitive intensity on relationship between the depen-
dent variable with Value. This is to estimate direct impact
that Value variable has on the dependent variables and also
the indirect impact Value variable has through competitive
intensity on the dependent variables.
In the implementation of the test using “paramed”
command of Stata, the two-stage regressions as described
in the methodology section is computed by the program
and results of estimates of direct, indirect and total impact
of Value variable on dependent variables is reported along
with their p-values. The estimate of the direct, indirect and
total impacts and their respective statistical significances
are of our interest. The results of the tests are presented in
Table 6.
The results of tests for Mediator Model improve our
understanding of the role competitive intensity plays as a
mediator in the relationship between Value variable and
dependent variables of Bid amount and Excess Bid. They
also confirm the results observed from the Base Model. In
case of the dependent variable being Bid amount, the results
of Mediator Model suggests that Value variable has statisti-
cally significant direct effect but no statistically significant
indirect effect. The total impact is statistically significant
on account of direct effect. This confirms the results from
the Base Model wherein we observe that Value variable has
statistically significant effect on Bid amount.
The results of Mediator Model when the dependent
variable is Excess Bid suggest that Value variable has no sta-
tistically significant effect on Excess Bid, neither directly
nor indirectly, and consequently no statistically significant
Table 5. Results of tests for Moderator Model
Dependent Variable →
Bid Amount Excess Bid
Estimate p-Value Estimate p-Value
Competitive intensity (N
i )148.31 0.008‡ 168.80 0.003‡
Value (Vi) 349.24 0.484 –345.47 0.489
Interaction (Moderation Effect) (Ni X Vi) 136.43 0.234 94.94 0.407
Number of observation =127
*p 0.10, † p 0.05, ‡ p 0.01.
increases by INR 142.67 per tonne while for every unit
increase in value variable of the coal blocks, the bid amount
increases by INR 859.29 per tonne. Both these estimates
have low p-values and, thus, are statistically significant.
The result of tests for Excess Bid shows that for
every unit increase in number of bidders, the Excess Bid
(bid amount over and above the comparable CIL prices)
increases by INR 164.87 per tonne while for every unit
increase in value variable of the coal blocks, the Excess bid
amount increases by INR 9.46 per tonne. The estimate for
the impact of competitive intensity on Excess Bid has low
p-value and, thus, is statistically significant. But the esti-
mate for the impact of Value variable on Excess Bid has
high p-value and, thus, is not statistically significant, indi-
cating that Value variable does impact Excess Bid in statisti-
cally significant way.
Result of Testing the Moderator Model
We ran the test for the Moderator Model, as in the Base
Model, with two scenarios of dependent variable being the
Bid amount and Excess Bid. This was to test the hypotheses
3 that is to estimate the moderating effect of competitive
intensity on relationship between the dependent variable
with value. In the implementation of the test, apart from
competitive intensity and value variables as independents,
we have an interaction term of both these independent
variables. The estimate of the coefficient of this interaction
term and its statistical significance are of our interest. The
results of the tests are presented in Table 5.
The results indicate that for Bid amount, the competi-
tive intensity does not appear to have a moderating effect
on the relationship between Value and Bid amount, even
though competitive intensity does impact the Bid amount
in statistically significant manner. The results indicate simi-
lar outcome for Excess Bid as well, the competitive inten-
sity does not appear to have a moderating effect on the
relationship between Value and Excess Bid, even though
competitive intensity does impact Excess Bid in statistically
significant manner. It is worthwhile to note that in both
cases of Bid amount and Excess Bid being the dependent
variable, when we test of moderation effect by including an
interaction term, the estimate for Value variable impact-
ing the dependent variables is seen to have no statistical
significance. That provides explanation for the outcome of
test that competitive intensity does not appear to impact
their relationship.
Result of Testing the Mediator Model
We ran the test for the Mediator Model, as in the Base and
Moderator Models, with two scenarios of dependent vari-
able being the Bid amount and Excess Bid. This was to test
the hypotheses 4 that is to estimate the mediating effect of
competitive intensity on relationship between the depen-
dent variable with Value. This is to estimate direct impact
that Value variable has on the dependent variables and also
the indirect impact Value variable has through competitive
intensity on the dependent variables.
In the implementation of the test using “paramed”
command of Stata, the two-stage regressions as described
in the methodology section is computed by the program
and results of estimates of direct, indirect and total impact
of Value variable on dependent variables is reported along
with their p-values. The estimate of the direct, indirect and
total impacts and their respective statistical significances
are of our interest. The results of the tests are presented in
Table 6.
The results of tests for Mediator Model improve our
understanding of the role competitive intensity plays as a
mediator in the relationship between Value variable and
dependent variables of Bid amount and Excess Bid. They
also confirm the results observed from the Base Model. In
case of the dependent variable being Bid amount, the results
of Mediator Model suggests that Value variable has statisti-
cally significant direct effect but no statistically significant
indirect effect. The total impact is statistically significant
on account of direct effect. This confirms the results from
the Base Model wherein we observe that Value variable has
statistically significant effect on Bid amount.
The results of Mediator Model when the dependent
variable is Excess Bid suggest that Value variable has no sta-
tistically significant effect on Excess Bid, neither directly
nor indirectly, and consequently no statistically significant
Table 5. Results of tests for Moderator Model
Dependent Variable →
Bid Amount Excess Bid
Estimate p-Value Estimate p-Value
Competitive intensity (N
i )148.31 0.008‡ 168.80 0.003‡
Value (Vi) 349.24 0.484 –345.47 0.489
Interaction (Moderation Effect) (Ni X Vi) 136.43 0.234 94.94 0.407
Number of observation =127
*p 0.10, † p 0.05, ‡ p 0.01.