Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 6

Sample Regression:

Ls lcons c ldisp lprice

Dependent Variable: LCONS


Method: Least Squares
Date: 04/13/19 Time: 11:01
Sample: 1985Q1 1994Q1
Included observations: 37

Variable Coefficient Std. Error t-Statistic Prob.

C 6.072816 1.633641 3.717351 0.0007


LDISP 0.295909 0.308278 0.959876 0.3439
LPRICE 0.032703 0.150500 0.217292 0.8293

R-squared 0.189750 Mean dependent var 9.024947


Adjusted R-squared 0.142088 S.D. dependent var 0.048217
S.E. of regression 0.044660 Akaike info criterion -3.301862
Sum squared resid 0.067814 Schwarz criterion -3.171247
Log likelihood 64.08445 Hannan-Quinn criter. -3.255814
F-statistic 3.981167 Durbin-Watson stat 0.452731
Prob(F-statistic) 0.027959

Graphically Detecting Auto Co-relation:


RES01
.12

.08

.04

.00

-.04

-.08
1985 1986 1987 1988 1989 1990 1991 1992 1993
0.2

0.0

-0.2

-0.4
-1
RES01 -0.6

-0.8

-1.0

-1.2
-.08 -.04 .00 .04 .08 .12

RES01

Both the graph show that the residual is serially correlated and particular positive serially correlated.

Durbin- Watson Test:

Breusch-Godfrey Serial Correlation LM Test:

F-statistic 13.30100 Prob. F(4,30) 0.0000


Obs*R-squared 23.65930 Prob. Chi-Square(4) 0.0001

Test Equation:
Dependent Variable: RESID
Method: Least Squares
Date: 04/13/19 Time: 14:53
Sample: 1985Q1 1994Q1
Included observations: 37
Presample missing value lagged residuals set to zero.

Variable Coefficient Std. Error t-Statistic Prob.

C -1.303296 1.093325 -1.192047 0.2426


LDISP 0.252705 0.210482 1.200604 0.2393
LPRICE -0.104858 0.102062 -1.027396 0.3124
RESID(-1) 0.793793 0.177562 4.470519 0.0001
RESID(-2) -0.317938 0.231859 -1.371256 0.1805
RESID(-3) 0.253458 0.232004 1.092472 0.2833
RESID(-4) 0.229214 0.212328 1.079526 0.2890

R-squared 0.639440 Mean dependent var 9.62E-16


Adjusted R-squared 0.567329 S.D. dependent var 0.043402
S.E. of regression 0.028549 Akaike info criterion -4.105744
Sum squared resid 0.024451 Schwarz criterion -3.800976
Log likelihood 82.95626 Hannan-Quinn criter. -3.998299
F-statistic 8.867334 Durbin-Watson stat 1.497774
Prob(F-statistic) 0.000013
Breusch-Godfrey Serial Correlation LM Test:

F-statistic 39.75375 Prob. F(1,33) 0.0000


Obs*R-squared 20.21736 Prob. Chi-Square(1) 0.0000

Test Equation:
Dependent Variable: RESID
Method: Least Squares
Date: 04/13/19 Time: 14:55
Sample: 1985Q1 1994Q1
Included observations: 37
Presample missing value lagged residuals set to zero.

Variable Coefficient Std. Error t-Statistic Prob.

C -1.011456 1.110740 -0.910614 0.3691


LDISP 0.223342 0.213700 1.045118 0.3036
LPRICE -0.108696 0.104319 -1.041959 0.3050
RESID(-1) 0.763825 0.121145 6.305057 0.0000

R-squared 0.546415 Mean dependent var 9.62E-16


Adjusted R-squared 0.505180 S.D. dependent var 0.043402
S.E. of regression 0.030530 Akaike info criterion -4.038381
Sum squared resid 0.030760 Schwarz criterion -3.864227
Log likelihood 78.71004 Hannan-Quinn criter. -3.976984
F-statistic 13.25125 Durbin-Watson stat 1.572165
Prob(F-statistic) 0.000008

Durbin’s h Test:
Ls lcons c ldisp lprice lcons(-1)

Dependent Variable: LCONS


Method: Least Squares
Date: 04/13/19 Time: 15:00
Sample (adjusted): 1985Q2 1994Q1
Included observations: 36 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

C -0.277651 1.237621 -0.224343 0.8239


LDISP 0.324522 0.183087 1.772500 0.0858
LPRICE -0.078810 0.092403 -0.852901 0.4001
LCONS(-1) 0.786842 0.106185 7.410118 0.0000

R-squared 0.729378 Mean dependent var 8.882058


Adjusted R-squared 0.704008 S.D. dependent var 0.048702
S.E. of regression 0.026496 Akaike info criterion -4.319173
Sum squared resid 0.022466 Schwarz criterion -4.143226
Log likelihood 81.74511 Hannan-Quinn criter. -4.257762
F-statistic 28.74876 Durbin-Watson stat 1.813960
Prob(F-statistic) 0.000000

H0= No auto correlation


H1= Auto correlation is present
Scalar h=1.062
So h < z-critical we reject Ho hypothesis and concluded that this model does not
suffer from serial correlation.

Breusch-Godfrey Serial Correlation LM Test:

F-statistic 0.348709 Prob. F(1,31) 0.5591


Obs*R-squared 0.400448 Prob. Chi-Square(1) 0.5269

Test Equation:
Dependent Variable: RESID
Method: Least Squares
Date: 04/13/19 Time: 15:18
Sample: 1985Q2 1994Q1
Included observations: 36
Presample missing value lagged residuals set to zero.

Variable Coefficient Std. Error t-Statistic Prob.

C 0.289098 1.342835 0.215289 0.8310


LDISP 0.005607 0.185223 0.030274 0.9760
LPRICE 0.006227 0.093951 0.066280 0.9476
LCONS(-1) -0.044342 0.130950 -0.338616 0.7372
RESID(-1) 0.129829 0.219857 0.590516 0.5591

R-squared 0.011124 Mean dependent var -8.41E-16


Adjusted R-squared -0.116473 S.D. dependent var 0.025335
S.E. of regression 0.026770 Akaike info criterion -4.274803
Sum squared resid 0.022216 Schwarz criterion -4.054870
Log likelihood 81.94645 Hannan-Quinn criter. -4.198040
F-statistic 0.087177 Durbin-Watson stat 1.906338
Prob(F-statistic) 0.985787

Resolving Auto correlation:


When p is known:
Ls lcons_star betal_star ldisp_star lprice_satr

Dependent Variable: LCONS_STAR


Method: Least Squares
Date: 04/13/19 Time: 15:32
Sample: 1985Q1 1994Q1
Included observations: 37

Variable Coefficient Std. Error t-Statistic Prob.

BETAL_STAR 5.314790 1.459586 3.641298 0.0009


LDISP_STAR 0.519566 0.523801 0.991916 0.3282
LPRICE_STAR -0.116058 0.399379 -0.290595 0.7731

R-squared 0.987039 Mean dependent var 11.27240


Adjusted R-squared 0.986276 S.D. dependent var 0.462404
S.E. of regression 0.054170 Akaike info criterion -2.915767
Sum squared resid 0.099770 Schwarz criterion -2.785152
Log likelihood 56.94169 Hannan-Quinn criter. -2.869719
Durbin-Watson stat 0.312150

When p is unknown:
Ls lcons c ldisp lprice ar(1)

Dependent Variable: LCONS


Method: ARMA Maximum Likelihood (BFGS)
Date: 04/13/19 Time: 15:37
Sample: 1985Q1 1994Q1
Included observations: 37
Convergence achieved after 10 iterations
Coefficient covariance computed using outer product of gradients

Variable Coefficient Std. Error t-Statistic Prob.

C 11.05276 2.498513 4.423733 0.0001


LDISP 0.535114 0.301010 1.777726 0.0850
LPRICE -0.780192 0.091513 -8.525506 0.0000
AR(1) 0.993225 0.031290 31.74308 0.0000
SIGMASQ 0.000182 5.15E-05 3.537304 0.0013

R-squared 0.919398 Mean dependent var 8.882772


Adjusted R-squared 0.909323 S.D. dependent var 0.048217
S.E. of regression 0.014519 Akaike info criterion -5.385230
Sum squared resid 0.006746 Schwarz criterion -5.167539
Log likelihood 104.6268 Hannan-Quinn criter. -5.308484
F-statistic 91.25343 Durbin-Watson stat 1.327833
Prob(F-statistic) 0.000000

Inverted AR Roots .99

ls lcons c ldisp lprice ar(1) ar(4)

Dependent Variable: LCONS


Method: ARMA Maximum Likelihood (BFGS)
Date: 04/13/19 Time: 15:38
Sample: 1985Q1 1994Q1
Included observations: 37
Convergence achieved after 17 iterations
Coefficient covariance computed using outer product of gradients

Variable Coefficient Std. Error t-Statistic Prob.

C 15.24886 3.219218 4.736821 0.0000


LDISP 0.131680 0.391332 0.336490 0.7388
LPRICE -0.854013 0.070877 -12.04917 0.0000
AR(1) 1.222176 0.064425 18.97058 0.0000
AR(4) -0.231433 0.064534 -3.586216 0.0011
SIGMASQ 0.000140 3.76E-05 3.728727 0.0008

R-squared 0.937939 Mean dependent var 8.882772


Adjusted R-squared 0.927929 S.D. dependent var 0.048217
S.E. of regression 0.012944 Akaike info criterion -5.554571
Sum squared resid 0.005194 Schwarz criterion -5.293341
Log likelihood 108.7596 Hannan-Quinn criter. -5.462475
F-statistic 93.70140 Durbin-Watson stat 2.115759
Prob(F-statistic) 0.000000

Inverted AR Roots .96 .87 -.30+.43i -.30-.43i

You might also like