Method: Least Squares Date: 06/11/14 Time: 22:10 Sample (adjusted): 1970 2012 Included observations: 43 after adjustments
Variable Coefficient Std. Error t-Statistic Prob.
C 14.40750 1.468117 9.813594 0.0000 LAID 0.298614 0.088261 3.383328 0.0016 LAIDV -0.182197 0.109598 -1.662410 0.1044 LFDI 0.216494 0.024982 8.666023 0.0000
R-squared 0.914844 Mean dependent var 24.74118 Adjusted R-squared 0.908294 S.D. dependent var 0.616956 S.E. of regression 0.186833 Akaike info criterion -0.428797 Sum squared resid 1.361353 Schwarz criterion -0.264965 Log likelihood 13.21914 Hannan-Quinn criter. -0.368381 F-statistic 139.6618 Durbin-Watson stat 1.266385 Prob(F-statistic) 0.000000
interpretation : the value of prob of LaidV is greater then 5 % which means there is no realation ship between Laidv and GDP where as there is relationship between laid, fdi and gdp t stat of laidv is less then 2 means no relation ship between the variables. R2 value is greater then 70 % means model is good.
Test Equation: Dependent Variable: RESID Method: Least Squares Date: 06/11/14 Time: 22:19 Sample: 1970 2012 Included observations: 43 Presample missing value lagged residuals set to zero.
R-squared 0.285637 Mean dependent var 0.031659 Adjusted R-squared 0.090811 S.D. dependent var 0.045657 S.E. of regression 0.043535 Akaike info criterion -3.230097 Sum squared resid 0.062544 Schwarz criterion -2.820516 Log likelihood 79.44709 Hannan-Quinn criter. -3.079056 F-statistic 1.466113 Durbin-Watson stat 1.746053 Prob(F-statistic) 0.201474
interpretation : Probability: probability of observed R squared is insignificance i-e more than 5% which means variance = constant and therefore there is no Heteroskedasticity.
Johansan test : Sample (adjusted): 1972 2012 Included observations: 41 after adjustments Trend assumption: No deterministic trend (restricted constant) Series: LAIDV LAID GDP FDI Lags interval (in first differences): 1 to 1
Unrestricted Cointegration Rank Test (Trace)
Hypothesized Trace 0.05 No. of CE(s) Eigenvalue Statistic Critical Value Prob.**
None * 0.746752 101.1550 54.07904 0.0000 At most 1 * 0.501151 44.84623 35.19275 0.0034 At most 2 0.190564 16.33269 20.26184 0.1594 At most 3 0.170507 7.664581 9.164546 0.0955
Trace test indicates 2 cointegrating eqn(s) at the 0.05 level * denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values
interpretation: the is less then 5% at most 1 and none hence there are two co integrating equations.
Multicolinearity :
Covariance Analysis: Ordinary Date: 06/11/14 Time: 22:28 Sample (adjusted): 1970 2012 Included observations: 43 after adjustments Balanced sample (listwise missing value deletion)
Covariance Correlation LFDI LGDP LAIDV LAID LFDI 4.039894 1.000000
LAIDV does not Granger Cause LFDI 41 0.85182 0.4351 LFDI does not Granger Cause LAIDV 0.12412 0.8837
LAID does not Granger Cause LFDI 41 0.70574 0.5004 LFDI does not Granger Cause LAID 3.35944 0.0460
LGDP does not Granger Cause LFDI 41 3.66039 0.0357 LFDI does not Granger Cause LGDP 4.38933 0.0197
LAID does not Granger Cause LAIDV 41 0.98681 0.3826 LAIDV does not Granger Cause LAID 416.413 1.E-25
LGDP does not Granger Cause LAIDV 41 1.03224 0.3665 LAIDV does not Granger Cause LGDP 0.26072 0.7719
LGDP does not Granger Cause LAID 41 4.19799 0.0230 LAID does not Granger Cause LGDP 1.22853 0.3047
interpretation : laidv causes Lfdi and LFDI causes LAIDV because there prob is greater then 5% laid granger cause Lfdi but lfdi does not granger cause laid. Laid granger cause laidv but laidv does not granger cause laid. Lgdp granger cause laidv and laidv granger cause lgdp. Lgdp does not granger cause laid and laid granger cause LGDP.
Lag Selection Criteria :
VAR Lag Order Selection Criteria Endogenous variables: LAID LAIDV LFDI LGDP Exogenous variables: C Date: 06/11/14 Time: 22:46 Sample: 1970 2013 Included observations: 36
* indicates lag order selected by the criterion LR: sequential modified LR test statistic (each test at 5% level) FPE: Final prediction error AIC: Akaike information criterion SC: Schwarz information criterion HQ: Hannan-Quinn information criterion
Interpretation: The Schwarz criteria indicates asterisk (*) on the 7th year lag, therefore we select 9th year lag as it is the best value- as revealed by SC criteria
VAR :
Vector Autoregression Estimates Date: 06/11/14 Time: 22:49 Sample (adjusted): 1972 2012 Included observations: 41 after adjustments Standard errors in ( ) & t-statistics in [ ]
Interpretation: Laidv at lag 1 has significant relationship with gdp. Lfdi at lag 1 has significant relationship with laid and lgdp . Lfdi at lag 2 has significant relationship with lgdp. Lgdp at lag 1 has significant relationship with laid.
Two stage least square method
Ct = 1 + 2 Yt
Dependent Variable: CONSP Method: Two-Stage Least Squares Date: 05/15/14 Time: 19:59 Sample: 1973 2011 Included observations: 39 Instrument list: GDP
Variable Coefficient Std. Error t-Statistic Prob.
C 3.79E+09 6.16E+08 6.154563 0.0000 GDP 0.676527 0.009369 72.21280 0.0000
R-squared 0.992955 Mean dependent var 4.34E+10 Adjusted R-squared 0.992764 S.D. dependent var 2.05E+10 S.E. of regression 1.75E+09 Sum squared resid 1.13E+20 F-statistic 5214.688 Durbin-Watson stat 1.156784 Prob(F-statistic) 0.000000 Second-Stage SSR 1.13E+20
INTERPRETATION: The coefficient of gdp (Yt) is positive and small, i.e. gdp has a significant positive relation with consumption (consp). The value of R2 is very high, indicating all the relevant variables are been in the equation.
It = v1 + v2 Yt + v3 Rt
Dependent Variable: INVST Method: Two-Stage Least Squares Date: 05/15/14 Time: 20:02 Sample (adjusted): 2004 2011 Included observations: 8 after adjustments Instrument list: GDP INT
Variable Coefficient Std. Error t-Statistic Prob.
C 2.04E+08 8.12E+09 0.025089 0.9810 GDP 0.027766 0.077311 0.359147 0.7342 INT 2.09E+08 2.83E+08 0.737377 0.4940
R-squared 0.099752 Mean dependent var 3.04E+09 Adjusted R-squared -0.260347 S.D. dependent var 1.80E+09 S.E. of regression 2.02E+09 Sum squared resid 2.04E+19 F-statistic 0.277013 Durbin-Watson stat 0.868142 Prob(F-statistic) 0.768963 Second-Stage SSR 2.04E+19
INTERPRETATION: The relationship of gdp and investment is positive but insignificant. Also the interest rate (int) has a positive relation, with small coefficient but insignificant. That is, investment is not been determined by the variables specified in the equation. The value of R2, hence, shows that the equation is missing some important variables; as R2 is extremely less.
Yt = Ct + It + Gt
Dependent Variable: GDP Method: Two-Stage Least Squares Date: 05/15/14 Time: 20:05 Sample: 1973 2011 Included observations: 39 Instrument list: CONSP INVST GOVT
R-squared 0.993683 Mean dependent var 5.86E+10 Adjusted R-squared 0.993142 S.D. dependent var 3.02E+10 S.E. of regression 2.50E+09 Sum squared resid 2.20E+20 F-statistic 1835.242 Durbin-Watson stat 0.909394 Prob(F-statistic) 0.000000 Second-Stage SSR 2.20E+20
INTERPRETATION: Consumption (Ct) has a positive significant relation with gdp, but the coefficient of consumption is small. However, investment and government expenditure have an insignificant relationship with gdp. The value of R2 is high, indicating that the equation is good as it has all important variables.
in this we can say that low volatility of aid is obsereved after another low volatility of aid. In the same way high volatility of aid is observed afther another high volatility of return. Hence there exist cluster volatility .
R-squared 0.703220 Mean dependent var 20.76694 Adjusted R-squared 0.653757 S.D. dependent var 0.592530 S.E. of regression 0.348659 Akaike info criterion 0.862315 Sum squared resid 4.376278 Schwarz criterion 1.149022 Log likelihood -11.53977 Hannan-Quinn criter. 0.968044 F-statistic 14.21699 Durbin-Watson stat 1.089475 Prob(F-statistic) 0.000000
@SQRT(GARCH) prob value is negative and insignificant. And tell us that is is not risky Variance equation RESID(-1)^2 = Arch we can see that the arch effect results show that it is positive and the value is greater then 5% means that the the volatality is not effected by Arch. That is it is not the internal cause of the volatility. And it is significant.
GARCH(-1) = Garch we can see that the value is positive and it is greater the 5% hence our value is insignificant and that is is also not an internal cause. it also shows us that the FDI volatility cannot effect the aid volatility. Serial autocorrelation :
Date: 05/28/14 Time: 17:50 Sample: 1970 2012 Included observations: 43
Autocorrelation Partial Correlation AC PAC Q-Stat Prob
considering the p value. we can not reject the null hypothesis and there is no serial correlation. there is arch effect but there is no serial correlation. hence that is the bad side of the model.
Two stage least square method
Ct = 1 + 2 Yt
Dependent Variable: CONSP Method: Two-Stage Least Squares Date: 05/15/14 Time: 19:59 Sample: 1973 2011 Included observations: 39 Instrument list: GDP
Variable Coefficient Std. Error t-Statistic Prob.
C 3.79E+09 6.16E+08 6.154563 0.0000 GDP 0.676527 0.009369 72.21280 0.0000
R-squared 0.992955 Mean dependent var 4.34E+10 Adjusted R-squared 0.992764 S.D. dependent var 2.05E+10 S.E. of regression 1.75E+09 Sum squared resid 1.13E+20 F-statistic 5214.688 Durbin-Watson stat 1.156784 Prob(F-statistic) 0.000000 Second-Stage SSR 1.13E+20
INTERPRETATION: The coefficient of gdp (Yt) is positive and small, i.e. gdp has a significant positive relation with consumption (consp). The value of R2 is very high, indicating all the relevant variables are been in the equation.
It = v1 + v2 Yt + v3 Rt
Dependent Variable: INVST Method: Two-Stage Least Squares Date: 05/15/14 Time: 20:02 Sample (adjusted): 2004 2011 Included observations: 8 after adjustments Instrument list: GDP INT
Variable Coefficient Std. Error t-Statistic Prob.
C 2.04E+08 8.12E+09 0.025089 0.9810 GDP 0.027766 0.077311 0.359147 0.7342 INT 2.09E+08 2.83E+08 0.737377 0.4940
R-squared 0.099752 Mean dependent var 3.04E+09 Adjusted R-squared -0.260347 S.D. dependent var 1.80E+09 S.E. of regression 2.02E+09 Sum squared resid 2.04E+19 F-statistic 0.277013 Durbin-Watson stat 0.868142 Prob(F-statistic) 0.768963 Second-Stage SSR 2.04E+19
INTERPRETATION: The relationship of gdp and investment is positive but insignificant. Also the interest rate (int) has a positive relation, with small coefficient but insignificant. That is, investment is not been determined by the variables specified in the equation. The value of R2, hence, shows that the equation is missing some important variables; as R2 is extremely less.
Yt = Ct + It + Gt
Dependent Variable: GDP Method: Two-Stage Least Squares Date: 05/15/14 Time: 20:05 Sample: 1973 2011 Included observations: 39 Instrument list: CONSP INVST GOVT
R-squared 0.993683 Mean dependent var 5.86E+10 Adjusted R-squared 0.993142 S.D. dependent var 3.02E+10 S.E. of regression 2.50E+09 Sum squared resid 2.20E+20 F-statistic 1835.242 Durbin-Watson stat 0.909394 Prob(F-statistic) 0.000000 Second-Stage SSR 2.20E+20
INTERPRETATION: Consumption (Ct) has a positive significant relation with gdp, but the coefficient of consumption is small. However, investment and government expenditure have an insignificant relationship with gdp. The value of R2 is high, indicating that the equation is good as it has all important variables.