HETROSKEDOSTICITY

ASSIGNMENT NO # 3

QAISAR SHAHZAD
BECF15M033
DEPARTMENT OF ECONOMICS
Sample regression:
Dependent Variable: PRICES
Method: Least Squares
Date: 03/22/19 Time: 16:27
Sample (adjusted): 1 88
Included observations: 88 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

C 336747.2 35864.82 9.389345 0.0000

ROOM -5864.877 4065.990 -1.442423 0.1529
SQFEETS -31.76323 15.13829 -2.098204 0.0389

R-squared 0.075256 Mean dependent var 245460.8

Adjusted R-squared 0.053497 S.D. dependent var 83585.01
S.E. of regression 81318.51 Akaike info criterion 25.48363
Sum squared resid 5.62E+11 Schwarz criterion 25.56809
Log likelihood -1118.280 Hannan-Quinn criter. 25.51766
F-statistic 3.458649 Durbin-Watson stat 1.893624
Prob(F-statistic) 0.035968

Breuch-Pagan Test:
Dependent Variable: UTSQ
Method: Least Squares
Date: 03/22/19 Time: 16:42
Sample (adjusted): 1 88
Included observations: 88 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

C 4.26E+09 3.01E+09 1.416817 0.1602

ROOM 6.93E+08 3.41E+08 2.033306 0.0451
SQFEETS -548276.6 1269481. -0.431890 0.6669

R-squared 0.047326 Mean dependent var 6.39E+09

Adjusted R-squared 0.024910 S.D. dependent var 6.91E+09
S.E. of regression 6.82E+09 Akaike info criterion 48.15741
Sum squared resid 3.95E+21 Schwarz criterion 48.24187
Log likelihood -2115.926 Hannan-Quinn criter. 48.19144
F-statistic 2.111272 Durbin-Watson stat 2.002492
Prob(F-statistic) 0.127390

H0 = Homodoskesticity
H1 = Hetroskedosticity

Chi = 5.99
Lm = 4.16

Interpretation:
If lm value is smaller than Chi, then we reject H0 and accept H1. It means that there is no
problem of hetroskedosticity. The above result show that LM (4.16) < Chi (5.9). So there is no
problem of hetroskedosticity in our data.

Glesjer Test:

Dependent Variable: ABSUT

Method: Least Squares
Date: 03/22/19 Time: 16:49
Sample (adjusted): 1 88
Included observations: 88 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

C 49552.94 18900.88 2.621726 0.0104

ROOM 4697.878 2142.791 2.192411 0.0311
SQFEETS -2.117574 7.977932 -0.265429 0.7913

R-squared 0.053673 Mean dependent var 67176.64

Adjusted R-squared 0.031406 S.D. dependent var 43544.37
S.E. of regression 42855.13 Akaike info criterion 24.20253
Sum squared resid 1.56E+11 Schwarz criterion 24.28699
Log likelihood -1061.912 Hannan-Quinn criter. 24.23656
F-statistic 2.410479 Durbin-Watson stat 1.939385
Prob(F-statistic) 0.095885

H0 = Homodoskesticity
H1 = Hetroskedosticity

Chi = 5.9
Lm = 4.723

Interpretation:
If lm value is smaller than Chi, then we reject H0 and accept H1. It means that there is no
problem of hetroskedosticity. The above result show that LM (4.723) < Chi (5.9). So there is
no problem of hetroskedosticity in our data.

Park Test:
Dependent Variable: LOG(UTSQ)
Method: Least Squares
Date: 03/22/19 Time: 17:25
Sample (adjusted): 1 88
Included observations: 88 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

C 17.60582 8.147750 2.160819 0.0335

LOG(ROOM) 1.226571 0.645670 1.899687 0.0609
 LOG(SQFEETS) 0.270185 1.079994 0.250173 0.8031

R-squared 0.042745 Mean dependent var 21.42222

Adjusted R-squared 0.020222 S.D. dependent var 2.620800
S.E. of regression 2.594166 Akaike info criterion 4.777904
Sum squared resid 572.0244 Schwarz criterion 4.862358
Log likelihood -207.2278 Hannan-Quinn criter. 4.811928
F-statistic 1.897800 Durbin-Watson stat 1.911591
Prob(F-statistic) 0.156195

H0 = Homodoskesticity
H1 = Hetroskedosticity

Chi = 5.99
Lm = 3.761560

Interpretation:
If lm value is smaller than Chi, then we reject H0 and accept H1. It means that there is no
problem of hetroskedosticity. The above result show that LM (4.723) < Chi (5.9). So there is
no problem of hetroskedosticity in our data.

Gold Feld Quant Test.

Dependent Variable: LOG(UTSQ)
Method: Least Squares
Date: 03/22/19 Time: 17:33
Sample (adjusted): 1 88
Included observations: 88 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

C 17.60582 8.147750 2.160819 0.0335

LOG(ROOM) 1.226571 0.645670 1.899687 0.0609
LOG(SQFEETS) 0.270185 1.079994 0.250173 0.8031

R-squared 0.042745 Mean dependent var 21.42222

Adjusted R-squared 0.020222 S.D. dependent var 2.620800
S.E. of regression 2.594166 Akaike info criterion 4.777904
Sum squared resid 572.0244 Schwarz criterion 4.862358
Log likelihood -207.2278 Hannan-Quinn criter. 4.811928
F-statistic 1.897800 Durbin-Watson stat 1.911591
Prob(F-statistic) 0.156195

Price = f(room)

Dependent Variable: PRICES

Method: Least Squares
Date: 03/22/19 Time: 17:36
Sample: 1 37
Included observations: 37
 Variable Coefficient Std. Error t-Statistic Prob.

C 249600.2 33327.28 7.489365 0.0000

ROOM -1196.569 6057.477 -0.197536 0.8446

R-squared 0.001114 Mean dependent var 243714.4

Adjusted R-squared -0.027426 S.D. dependent var 89592.75
S.E. of regression 90813.02 Akaike info criterion 25.72353
Sum squared resid 2.89E+11 Schwarz criterion 25.81061
Log likelihood -473.8853 Hannan-Quinn criter. 25.75423
F-statistic 0.039020 Durbin-Watson stat 1.844354
Prob(F-statistic) 0.844551

Dependent Variable: PRICES

Method: Least Squares
Date: 03/22/19 Time: 17:40
Sample: 51 88
Included observations: 38

Variable Coefficient Std. Error t-Statistic Prob.

C 317391.6 33520.07 9.468702 0.0000

ROOM -14053.16 6521.236 -2.154985 0.0379

R-squared 0.114260 Mean dependent var 250824.0

Adjusted R-squared 0.089656 S.D. dependent var 84090.06
S.E. of regression 80231.99 Akaike info criterion 25.47443
Sum squared resid 2.32E+11 Schwarz criterion 25.56062
Log likelihood -482.0141 Hannan-Quinn criter. 25.50509
F-statistic 4.643960 Durbin-Watson stat 2.232493
Prob(F-statistic) 0.037927

H0 = Homodoskesticity
H1 = Hetroskedosticity

F-Crit = 1.72
F-Calculate = 1.8

Interpretation:

If F calculated value is greater than F critical, then we reject H0 and accept H1. It means that
there is no problem of hetroskedosticity. The above result show that F calculated > F critical.
So there is no problem of hetroskedosticity in our data.

White Test:
Heteroskedasticity Test: White

F-statistic 1.834218 Prob. F(5,82) 0.1152

Obs*R-squared 8.852102 Prob. Chi-Square(5) 0.1151
Scaled explained SS 4.772307 Prob. Chi-Square(5) 0.4443

Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 03/22/19 Time: 18:00
Sample: 1 88
Included observations: 88

Variable Coefficient Std. Error t-Statistic Prob.

C -1.66E+10 1.06E+10 -1.564124 0.1216

ROOM^2 -83918258 1.62E+08 -0.517505 0.6062
ROOM*SQFEETS -920799.0 627647.6 -1.467064 0.1462
ROOM 3.49E+09 2.03E+09 1.717107 0.0897
SQFEETS^2 -1838.836 1651.582 -1.113379 0.2688
SQFEETS 12787541 8104616. 1.577810 0.1185

R-squared 0.100592 Mean dependent var 6.39E+09

Adjusted R-squared 0.045750 S.D. dependent var 6.91E+09
S.E. of regression 6.75E+09 Akaike info criterion 48.16806
Sum squared resid 3.73E+21 Schwarz criterion 48.33697
Log likelihood -2113.395 Hannan-Quinn criter. 48.23611
F-statistic 1.834218 Durbin-Watson stat 2.123082
Prob(F-statistic) 0.115197

H0 = Homodoskesticity
H1 = Hetroskedosticity

Chi = 5.99
Lm = 4.761560

Interpretation:
If lm value is smaller than Chi, then we reject H0 and accept H1. It means that there is no
problem of hetroskedosticity. The above result show that LM (4.761) < Chi (5.9). So there is
no problem of hetroskedosticity in our data.

ARCH Test:
Heteroskedasticity Test: ARCH

F-statistic 0.088483 Prob. F(1,85) 0.7668

Obs*R-squared 0.090471 Prob. Chi-Square(1) 0.7636

Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 03/22/19 Time: 18:03
Sample (adjusted): 2 88
Included observations: 87 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

C 6.24E+09 1.02E+09 6.121398 0.0000

RESID^2(-1) 0.032172 0.108154 0.297461 0.7668

R-squared 0.001040 Mean dependent var 6.44E+09

Adjusted R-squared -0.010713 S.D. dependent var 6.93E+09
S.E. of regression 6.96E+09 Akaike info criterion 48.18830
Sum squared resid 4.12E+21 Schwarz criterion 48.24499
Log likelihood -2094.191 Hannan-Quinn criter. 48.21113
F-statistic 0.088483 Durbin-Watson stat 1.894793
Prob(F-statistic) 0.766840

H0 = Homodoskesticity
H1 = Hetroskedosticity

Chi = 5.99
Lm = 3.761560

Interpretation:
If lm value is smaller than Chi, then we reject H0 and accept H1. It means that there is no
problem of hetroskedosticity. The above result show that LM (3.761) < Chi (5.9). So there is
no problem of hetroskedosticity in our data.