In this chapter, the regression
result will be subjected to various tests under econometric research
methodology and with that an analysis will be made.
PRESENTATION OF RESULTS - EVALUATION OF RESULTS
Evaluation based on Economic criteria (A priori
expectation)
All the estimated parameters should
have the right signs. As it is explained, economic and a priori criteria are
determined by principles of economic theory and refer to signs and magnitude of
the parameters in the model, as they are compared with out stated expectations
in the previous chapter. In this stage, we check whether the parameters
estimated in the model conforms to the a prior expectation in the theory in
terms of the size, signs and magnitude.
Expected and obtained signs of parameters
variables
|
Expected
sign
|
Obtained
sign
|
Remarks
|
RGDP
|
>0(Positive)
|
>0(Positive)
|
Conforms
|
CAD
|
>0(Positive)
|
>0(Positive)
|
Conforms
|
REER
|
<0(Negative)
|
<0(Negative)
|
Conforms
|
RIR
|
>0(Positive)
|
>0(Positive)
|
Conforms
|
From the table, it
can be seen that the negative relationship between BD and FGCE was not
established because of inaccurate measure of calculating federal government
capital expenditure.
Evaluation based on statistically Criteria (First Order) Test
These tests are determined by
Statistical theory and aim at evaluating reliability of the estimates and
parameters of the model (Koutsoyiannis, 1977).
Coefficient
of Determination (R2)
In our model, R2 = which implies that approximately % of
the variation in the dependent variable (BD) is caused by the variation in the
explanatory variables (Real gross domestic product, current account deficit,
real exchange rate, federal government capital expenditure and real interest
rate). Judging by the size of the coefficient of determination (R2), % shows a good fit for the model, meaning
that % variation is explained in the
model while % variations in the model attributes
to other factors not included in the model.
The
t-statistics
This form of test involves comparing
the estimated t-statistic with its tabular value at a chosen level of
significance. In this section, we use t-static to test the individual
significance of the parameter.
Hypothesis:
H0: β = 0 – The parameter is statistically
insignificant
H0: β ≠
0 - The parameter is statistically significant.
Decision
rule:
Reject Ho if tcal > ttab, accept if otherwise.
t0.025 =
Table
4.2 Statistical Significance Of The
Parameters
variables
|
T-statistics
|
Critical
value
|
conclusion
|
RGDP
|
Statistically
insignificant
|
||
CAD
|
Conforms
|
||
REER
|
Conforms
|
||
RIR
|
Conforms
|
F-ratio
Test
This test involves testing the
overall significance of the regression results as against individual
significance of the regression. This test can be said to be a join hypothesis
test employing Analysis of Varaicne (ANOVA) Gujarati (1995:245-246).
The
F-calculated value is
86.926 while the F-tabulated
value is 3.37 at 5% level of significance. Since the F-calculated
value is greater than the F-tabulated
value, we conclude that the entire regression plane is statistically
significant. This means that the joint influence of the explanatory variables
(RGEE and CGEE) on the dependent variable (RGDP) is statistically significant.
Durbin
Watson statistics: It is used
to test for the presence of positive first order serial correlation. The
computed DW is 0.737. At 5% level of significance with two explanatory
variables and 29 observations, the tabulated DW for dL and du are 1.270 and
1.563 respectively. The value of DW is less than the lower limit. Therefore, we
conclude that there is evidence of positive first order serial correlation.
TEST OF HYPOTHESIS
The
researcher examines the impact of budget deficit in Nigeria’s economic growth.
With respect to this, the null and alternative hypotheses are stated as
follows;
H0: There is no significant
relationship between budget deficit and economic growth in Nigeria.
H1: There is significant
relationship between budget deficit and economic growth in Nigeria.
F-test is employed in testing the
hypothesis. This test will help to capture the joint influence of the explanatory variables on the
dependent variable.
Decision Rule
If F-cal > F-tab, reject the null
hypothesis otherwise accept the null hypothesis. Using 5% level of significance
at 2 and 26 degrees of freedom, the tabulated F- value is 3.37 while calculated
F-value is 86.926. Since the calculated F-value is greater than the tabulated
F-value at 5% level of significance; we reject the null hypothesis and conclude
that budget deficit has significant impact on economic growth of Nigeria.
POLICY
IMPLICATION OF THE RESEARCH FINDINGS
So far, I have critically analyzed
the research findings. However, it is important at this level to state the
economic or policy implications of our findings.
The
statistical significant positive coefficient of real gross domestic product is
consistent with the view that as Nigerian RGDP increases, the higher the
tendency to increase the level of our budget deficit.
Also, the
statistical significant positive coefficient of current account deficit is
consistent with the view of the existence of the twin deficits in Nigeria.
Thus, for the level of budget deficit to be curtailed in Nigeria, policy instruments
should be applied to ensure that exports must exceed imports into Nigeria.
The real exchange
rate shows a statistically significant negative coefficient showing that a fall
in real exchange rate (exchange rate appreciation) is a better option for Nigerian
to curtail its level of budget deficit.
Also, the
federal government capital expenditure shows a statistically significant negative
coefficient revealing the fact that in Nigeria, government actually runs a budget
deficit almost every year yet, its expenditure on capital goods is not
improving.
Real
interest rate is statically insignificant and shows a negative coefficient
showing that increase in budget deficit lower interest rate which leads to fall
in investment. The fall in investment further leads to increase imports which
may virtually reduce exports and further create current account deficits.
The statistically significant variables
affecting budget deficit in Nigeria such as current account deficit, real exchange
rate, federal government capital expenditure and real gross domestic product
should be put on check regarding their positive or negative relationship with budget
deficit in Nigeria. Thus, effective manipulation of these variables can help
check the changes in the level of budget deficit in Nigeria.
READ RELATED TOPICS ON
DEFICIT BUDGET
SUMMARY, CONCLUSION AND
RECOMMENDATION OF BUDGET DEFICIT
METHODOLOGY OF BUDGET
DEFICIT (TECHNIQUES, MODEL, STATISTICAL DATA