DATA
ON GDP GROWTH RATE, EXCHANGE RATE AND INFLATION
|
YEAR
|
EXCHANGE
RATE
|
GDP
RATE
|
INFLATION
RATE
|
|
2017
|
305.3
|
0.8
|
16.5
|
|
2016
|
253.5
|
-1.6
|
15.7
|
|
2015
|
192.4
|
2.7
|
9.0
|
|
2014
|
158.6
|
6.3
|
8.1
|
|
2013
|
157.3
|
5.4
|
8.5
|
|
2012
|
157.5
|
4.3
|
12.2
|
|
2011
|
153.9
|
4.9
|
10.8
|
|
2010
|
150.3
|
7.8
|
13.7
|
|
2009
|
148.9
|
6.9
|
11.5
|
|
2008
|
118.5
|
6.3
|
11.6
|
|
2007
|
125.8
|
6.8
|
5.4
|
|
2006
|
128.7
|
8.2
|
8.2
|
|
2005
|
131.3
|
3.4
|
17.9
|
|
2004
|
132.9
|
33.7
|
15.0
|
|
2003
|
129.2
|
10.4
|
14.0
|
|
2002
|
120.6
|
3.8
|
12.9
|
|
2001
|
111.2
|
4.4
|
18.9
|
|
2000
|
101.7
|
5.3
|
6.9
|
|
1999
|
92.3
|
0.5
|
6.6
|
|
1998
|
21.9
|
2.7
|
10.0
|
Source: Central Bank of Nigeria (CBN)
statistical bulletin, 2016
MULTIPLE REGRESSIONS
|
Variables
Entered/Removeda
|
|||
|
Model
|
Variables
Entered
|
Variables
Removed
|
Method
|
|
1
|
INFLATION RATE, EXCHANGE RATEb
|
.
|
Enter
|
|
a. Dependent Variable: GROWTH RATE
|
|||
|
b. All requested variables entered.
|
|||
The above table in the result
output tells us the variables in our analysis, tells us that inflation rate and
exchange rate are useful for the prediction
|
Model
Summaryb
|
||||||||
|
Model
|
R
|
R Square
|
Adjusted R
Square
|
Std. Error
of the Estimate
|
Change Statistics
|
|||
|
R Square
Change
|
F Change
|
df1
|
||||||
|
1
|
.802a
|
.643
|
.564
|
2.12689
|
.643
|
8.115
|
2
|
|
|
a. Predictors: (Constant), INFLATION RATE, EXCHANGE RATE
|
||||||||
|
b. Dependent Variable: GROWTH RATE
|
||||||||
The above table is
the multiple linear regression model and overall fit statistics. We find that
the adjusted R2 of our model is 564 with R2 = 643. This means that the linear
regression explains 64.3% of the variance in the data.
|
ANOVAa
|
||||||
|
Model
|
Sum of
Squares
|
df
|
Mean
Square
|
F
|
Sig.
|
|
|
1
|
Regression
|
73.417
|
2
|
36.709
|
8.115
|
.010b
|
|
Residual
|
40.713
|
9
|
4.524
|
|
|
|
|
Total
|
114.130
|
11
|
|
|
|
|
|
a. Dependent Variable: GROWTH RATE
|
||||||
|
b. Predictors: (Constant), INFLATION RATE, EXCHANGE RATE
|
||||||
The above output table is the
f-test. The linear regression f-test has the null hypothesis that the model
explain zero variance in the dependent variable (in other words R2 =
0). The f-test is highly significant, thus we can assume that the model explain
a significant amount of variance in inflation rate.
|
Coefficientsa
|
||||||||
|
Model
|
Unstandardized
Coefficients
|
Standardized
Coefficients
|
t
|
Sig.
|
||||
|
B
|
Std. Error
|
Beta
|
VIF
|
|||||
|
1
|
(Constant)
|
16.907
|
3.091
|
|
5.470
|
.000
|
|
|
|
EXCHANGE RATE
|
-.072
|
.018
|
-.815
|
-3.917
|
.004
|
1.092
|
||
|
INFLATION RATE
|
.045
|
.190
|
.049
|
.238
|
.817
|
1.092
|
||
|
a. Dependent Variable: GROWTH RATE
|
||||||||
If we would have forced all
variables into the linear regression model, we would have seen a slightly
higher R2 and adjusted R2 (802 and 564 respectively).
|
Collinearity
Diagnosticsa
|
||||||
|
Model
|
Dimension
|
Eigenvalue
|
Condition
Index
|
Variance
Proportions
|
||
|
(Constant)
|
EXCHANGE
RATE
|
INFLATION
RATE
|
||||
|
1
|
1
|
2.923
|
1.000
|
.00
|
.00
|
.01
|
|
2
|
.053
|
7.397
|
.09
|
.18
|
.97
|
|
|
3
|
.024
|
11.092
|
.91
|
.82
|
.02
|
|
|
a. Dependent Variable: GROWTH RATE
|
||||||
|
Residuals Statisticsa
|
|||||
|
|
Minimum
|
Maximum
|
Mean
|
Std.
Deviation
|
N
|
|
Predicted Value
|
-.6506
|
8.8918
|
6.1367
|
2.58347
|
12
|
|
Residual
|
-1.89871
|
5.05508
|
.00000
|
1.92385
|
12
|
|
Std. Predicted Value
|
-2.627
|
1.066
|
.000
|
1.000
|
12
|
|
Std. Residual
|
-.893
|
2.377
|
.000
|
.905
|
12
|
|
a. Dependent Variable: GROWTH RATE
|
|||||
Charts
The p-p plot for checking for
normality of residuals. The plot shows that the point generally follows the
normal (diagonal) line with no strong deviations.
Please note: The analysis was done using s0pss
No comments:
Post a Comment