You are on page 1of 7

STATISTICAL ANALYSIS

Hypothesis:

Fiscal deficit is the major cause of inflation

Data Analysis

We convert all the data in the log form and do the analysis. Initially we check whether all the data are
stationary and this is done with the help of the unit root test.

Augemented Dickey Fuller Test for Gross Domestic Product

Null Hypothesis: LOGGDP has a unit root


Exogenous: Constant
Lag Length: 0 (Automatic based on SIC, MAXLAG=9)

t-Statistic   Prob.*

Augmented Dickey-Fuller test statistic -2.396512  0.1496


Test critical values: 1% level -3.621023
5% level -2.943427
10% level -2.610263

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation


Dependent Variable: D(LOGGDP)
Method: Least Squares
Date: 12/12/10 Time: 02:38
Sample (adjusted): 1972 2009
Included observations: 37 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

LOGGDP(-1) -0.267156 0.111477 -2.396512 0.0220


C 0.455056 0.185790 2.449303 0.0195

R-squared 0.140962     Mean dependent var 0.017384


Adjusted R-squared 0.116418     S.D. dependent var 0.220841
S.E. of regression 0.207589     Akaike info criterion -0.253978
Sum squared resid 1.508257     Schwarz criterion -0.166902
Log likelihood 6.698600     F-statistic 5.743270
Durbin-Watson stat 1.960543     Prob(F-statistic) 0.022029

Results:

There presence of unit root in the Data. Hence we need to take first differential of the GDP data.

First Differential

Null Hypothesis: D(LOGGDP) has a unit root


Exogenous: Constant
Lag Length: 1 (Automatic based on SIC, MAXLAG=9)

t-Statistic   Prob.*

Augmented Dickey-Fuller test statistic -4.959655  0.0003


Test critical values: 1% level -3.632900
5% level -2.948404
10% level -2.612874

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation


Dependent Variable: D(LOGGDP,2)
Method: Least Squares
Date: 12/12/10 Time: 02:41
Sample (adjusted): 1974 2009
Included observations: 35 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

D(LOGGDP(-1)) -1.453215 0.293007 -4.959655 0.0000


D(LOGGDP(-1),2) 0.327525 0.213658 1.532942 0.1351
C 0.025274 0.035798 0.706005 0.4853

R-squared 0.597303     Mean dependent var 0.014786


Adjusted R-squared 0.572135     S.D. dependent var 0.323024
S.E. of regression 0.211295     Akaike info criterion -0.189309
Sum squared resid 1.428656     Schwarz criterion -0.055993
Log likelihood 6.312900     F-statistic 23.73215
Durbin-Watson stat 1.894830     Prob(F-statistic) 0.000000

Resutls: The GDP data is stationary at the first differential.


Stationarity check for the Inflation

Null Hypothesis: LOGCPI has a unit root


Exogenous: Constant
Lag Length: 0 (Automatic based on SIC, MAXLAG=9)

t-Statistic   Prob.*

Augmented Dickey-Fuller test statistic -4.509021  0.0009


Test critical values: 1% level -3.621023
5% level -2.943427
10% level -2.610263

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation


Dependent Variable: D(LOGCPI)
Method: Least Squares
Date: 12/12/10 Time: 02:43
Sample (adjusted): 1972 2009
Included observations: 37 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

LOGCPI(-1) -0.716443 0.158891 -4.509021 0.0001


C 1.456640 0.323483 4.502979 0.0001

R-squared 0.367446     Mean dependent var 0.036248


Adjusted R-squared 0.349373     S.D. dependent var 0.554647
S.E. of regression 0.447386     Akaike info criterion 1.281748
Sum squared resid 7.005394     Schwarz criterion 1.368824
Log likelihood -21.71233     F-statistic 20.33127
Durbin-Watson stat 1.957472     Prob(F-statistic) 0.000070

Results: We find that the data is fairly stationary.

Stationarity check for the GDP data

Null Hypothesis: LOGGDP has a unit root


Exogenous: Constant
Lag Length: 0 (Automatic based on SIC, MAXLAG=9)

t-Statistic   Prob.*
Augmented Dickey-Fuller test statistic  2.941735  1.0000
Test critical values: 1% level -3.621023
5% level -2.943427
10% level -2.610263

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation


Dependent Variable: D(LOGGDP)
Method: Least Squares
Date: 12/12/10 Time: 02:45
Sample (adjusted): 1972 2009
Included observations: 37 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

LOGGDP(-1) 0.029134 0.009904 2.941735 0.0058


C -0.347660 0.138159 -2.516366 0.0166

R-squared 0.198237     Mean dependent var 0.058393


Adjusted R-squared 0.175330     S.D. dependent var 0.039798
S.E. of regression 0.036141     Akaike info criterion -3.750231
Sum squared resid 0.045716     Schwarz criterion -3.663155
Log likelihood 71.37928     F-statistic 8.653804
Durbin-Watson stat 2.058158     Prob(F-statistic) 0.005755

Results: The data is not stationary according to the results. We need to take the first differential of the
data.

Stationary check of GDP at first differntial

Null Hypothesis: D(LOGGDP) has a unit root


Exogenous: Constant
Lag Length: 0 (Automatic based on SIC, MAXLAG=9)

t-Statistic   Prob.*

Augmented Dickey-Fuller test statistic -5.116475  0.0002


Test critical values: 1% level -3.626784
5% level -2.945842
10% level -2.611531

*MacKinnon (1996) one-sided p-values.


Augmented Dickey-Fuller Test Equation
Dependent Variable: D(LOGGDP,2)
Method: Least Squares
Date: 12/12/10 Time: 02:47
Sample (adjusted): 1973 2009
Included observations: 36 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

D(LOGGDP(-1)) -0.834784 0.163156 -5.116475 0.0000


C 0.050486 0.011503 4.388868 0.0001

R-squared 0.435012     Mean dependent var 0.001894


Adjusted R-squared 0.418395     S.D. dependent var 0.051066
S.E. of regression 0.038944     Akaike info criterion -3.599411
Sum squared resid 0.051567     Schwarz criterion -3.511437
Log likelihood 66.78939     F-statistic 26.17832
Durbin-Watson stat 2.060827     Prob(F-statistic) 0.000012

Results : The data is fairly stationary.

Relationship between inflation and fiscal deficit

Dependent Variable: LOGCPI


Method: Least Squares
Date: 12/12/10 Time: 02:53
Sample: 1971 2009
Included observations: 38
LOGCPI = C(1) + C(2)*LOGDEFICIT

Coefficient Std. Error t-Statistic Prob.  

C(1) 1.656379 0.422476 3.920645 0.0004


C(2) 0.207607 0.252530 0.822111 0.4164

R-squared 0.018428     Mean dependent var 1.997887


Adjusted R-squared -0.008838     S.D. dependent var 0.472437
S.E. of regression 0.474520     Akaike info criterion 1.398171
Sum squared resid 8.106104     Schwarz criterion 1.484360
Log likelihood -24.56526     Durbin-Watson stat 1.411307
Output

Estimation Equation:
=====================
LOGCPI = C(1) + C(2)*LOGDEFICIT

Substituted Coefficients:
=====================
LOGCPI = 1.65637853 + 0.2076072895*LOGDEFICIT

Results

 We found that the Inflation rate is positively related with the Fiscal Deficit
 The model is not good as the R square Value is considerably low in value.
 Presence of high amount of residuals suggest many other factors to affect the equation

Relation between Fiscal deficit, GDP and Inflation

Dependent Variable: LOGCPI


Method: Least Squares
Date: 12/12/10 Time: 03:08
Sample: 1971 2009
Included observations: 38
LOGCPI = C(1) + C(2)*LOGDEFICIT + C(3)*LOGGDP

Coefficient Std. Error t-Statistic Prob.  

C(1) 3.446144 1.741930 1.978348 0.0558


C(2) 0.222647 0.252505 0.881754 0.3839
C(3) -0.129869 0.122637 -1.058971 0.2969

R-squared 0.048902     Mean dependent var 1.997887


Adjusted R-squared -0.005447     S.D. dependent var 0.472437
S.E. of regression 0.473722     Akaike info criterion 1.419265
Sum squared resid 7.854443     Schwarz criterion 1.548548
Log likelihood -23.96604     Durbin-Watson stat 1.458892

Estimation Equation:
=====================
LOGCPI = C(1) + C(2)*LOGDEFICIT + C(3)*LOGGDP

Substituted Coefficients:
=====================
LOGCPI = 3.446144073 + 0.2226469602*LOGDEFICIT - 0.1298686663*LOGGDP
Final Results

We find that there is low relationship between the GDP, Inflation and Fiscal Deficit

Fiscal Deficit cannot be called as a cause of inflation. Hence we reject the hypothesis.

You might also like