In: Economics
Based on the EViews result in Table 1, answer the following questions.
Table 1: Estimated Regression Result
Dependent Variable: LGDP |
||||
Method: Least Squares |
||||
Sample: 2000Q1 2012Q4 |
||||
Included observations: 52 |
||||
Variable |
Coefficient |
Std. Error |
t-Statistic |
Prob. |
LM2 |
0.796910 |
0.032383 |
24.60868 |
0.0000 |
LREER |
-0.060125 |
0.065220 |
-0.921892 |
0.3613 |
LSDR |
0.051745 |
0.074618 |
0.693465 |
0.4914 |
LTBR |
0.029174 |
0.007715 |
3.781436 |
0.0004 |
C |
2.795406 |
0.427355 |
6.541183 |
0.0000 |
R-squared |
0.986591 |
Mean dependent var |
13.28409 |
|
Adjusted R-squared |
0.985450 |
S.D. dependent var |
0.429315 |
|
S.E. of regression |
0.051786 |
Akaike info criterion |
-2.992187 |
|
Sum squared resid |
0.126044 |
Schwarz criterion |
-2.804568 |
|
Log likelihood |
82.79686 |
Hannan-Quinn criter. |
-2.920258 |
|
F-statistic |
864.5234 |
Durbin-Watson stat |
0.777639 |
|
Prob(F-statistic) |
0.000000 |
|||
a.
The estimated result is:
LGDP = 2.795406 + 0.796910*LM2 - 0.060125*LREER + 0.051745*LSDR + 0.029174*LTBR + e
b.
LTBR and LM2 are statistically significant at a 5% level of significance as the actual p-value of the respective t-statistic is less than the critical p-value of 0.05
LREER and LSDR are statistically not significant at a 5% level of significance as the actual p-value of the respective t-statistic is greater than the critical p-value of 0.05
c.
LM2 = log of money supply M2
LSDR = log of Special Drawing Rights (SDR)
LREER = log of Real Effective Exchange Rate
LTBR = log of Treasury-bill rate
The explanatory variables LTBR and LM2 are statistically significant variables in the explanation of variation in the dependent variable LGDP, whereas, LREER and LSDR are not.
Everything else remaining constant, 1% point increase in the M2 money supply would lead to 0.796910% point increase in the GDP
Everything else remaining constant, a 1% point increase in the TBR would lead to 0.029174% point increase in the GDP
d.
The Adjusted R-square is very high, which is 98.545%, but 2 of the 4 coefficients are statistically not significant in explaining the variations in the dependent variable GDP. The OLS regression model is likely to suffer from the problem of multi-collinearity.
The time-series data is used to build the regression model. There is likely a chance, the regression model suffers from the problem of serial-correlation.