Normality

Dependent Variable: LOG___CRUDE_OIL_

Method: Least Squares
Date: 09/15/24 Time: 09:59
Sample (adjusted): 2 2472
Included observations: 2469 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

LOG___PERSISTENT__SYSTEMS_ -0.043699 0.027097 -1.612678 0.1069

C -0.000508 0.000579 -0.878858 0.3796

R-squared 0.001053 Mean dependent var -0.000461

Adjusted R-squared 0.000648 S.D. dependent var 0.028719
S.E. of regression 0.028710 Akaike info criterion -4.262359
Sum squared resid 2.033407 Schwarz criterion -4.257651
Log likelihood 5263.882 Hannan-Quinn criter. -4.260649
F-statistic 2.600730 Durbin-Watson stat 1.905294
Prob(F-statistic) 0.106942

1,200
Series: Residual s
Sample 2 2472
1,000
Observations 2469

800 Mean 5.87e-19

Median -0.001253
600 Maximum 0.282066
Minimum -0.319297
400 Std. Dev. 0.028704
Skewness -0.106231
200 Kurtosis 26.89722

0 Jarque-Bera 58754.22
-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 Probability 0.000000
AUTOCORRELATION
Date: 09/15/24 Time: 10:14
Sample (adjusted): 2 2472
Included observations: 2471 after adjustments
Autocorrelation Partial Correlation AC PAC Q-Stat Prob

1 0.046 0.046 5.2556 0.022

2 0.004 0.002 5.2890 0.071
3 0.027 0.027 7.1475 0.067
4 0.017 0.014 7.8317 0.098
5 0.006 0.004 7.9131 0.161
6 -0.011 -0.012 8.2036 0.224
7 0.000 0.000 8.2036 0.315
8 0.020 0.020 9.2199 0.324
9 0.040 0.039 13.201 0.154
10 -0.006 -0.010 13.302 0.207
11 -0.017 -0.018 14.036 0.231
12 0.020 0.018 14.986 0.242
13 -0.004 -0.007 15.025 0.306
14 -0.033 -0.032 17.792 0.216
15 0.001 0.005 17.796 0.274
16 0.036 0.036 21.045 0.177
17 0.038 0.035 24.630 0.103
18 0.007 0.003 24.738 0.132
19 0.033 0.032 27.383 0.096
20 0.013 0.007 27.773 0.115
21 -0.018 -0.022 28.540 0.126
22 0.015 0.017 29.107 0.142
23 -0.016 -0.014 29.712 0.158
24 -0.032 -0.034 32.339 0.119
25 0.007 0.005 32.456 0.145
26 0.016 0.016 33.107 0.159
27 -0.008 -0.008 33.264 0.188
28 0.001 -0.001 33.267 0.226
29 -0.031 -0.032 35.604 0.185
30 -0.018 -0.011 36.391 0.196
31 -0.024 -0.023 37.892 0.184
32 0.013 0.018 38.292 0.205
33 -0.029 -0.026 40.439 0.175
34 -0.004 -0.004 40.485 0.206
35 0.048 0.043 46.201 0.098
36 0.029 0.027 48.307 0.082

600
Series: LOG__TATA_STEEL_
Sampl e 1 2473
500
Observations 2471

400 Mean -0.000483

Median -0.000500
300 Maximum 0.141000
Minimum -0.127800
200 Std. Dev. 0.023299
Skewness 0.184568
100 Kurtosis 6.101670

Jarque-Bera 1004.525
0
-0.10 -0.05 0.00 0.05 0.10 0.15 Probability 0.000000
STATIONARITY
Null Hypothesis: LOG___PERSISTENT__SYSTEMS_ has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=26)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -47.43455 0.0001

Test critical values: 1% level -3.432803
5% level -2.862510
10% level -2.567331

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

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(LOG___PERSISTENT__SYSTEMS_)
Method: Least Squares
Date: 09/15/24 Time: 10:20
Sample (adjusted): 3 2472
Included observations: 2470 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

LOG___PERSISTENT__SYSTEMS_(-1) -0.953898 0.020110 -47.43455 0.0000

C -0.001053 0.000429 -2.452954 0.0142

R-squared 0.476901 Mean dependent var 3.40E-06

Adjusted R-squared 0.476689 S.D. dependent var 0.029466
S.E. of regression 0.021316 Akaike info criterion -4.857906
Sum squared resid 1.121392 Schwarz criterion -4.853200
Log likelihood 6001.514 Hannan-Quinn criter. -4.856197
F-statistic 2250.036 Durbin-Watson stat 1.999619
Prob(F-statistic) 0.000000

600
Series: LOG___PERSISTENT__SYSTEM
Sample 1 2473
500
Observations 2471

400
Mean -0.001101
Median -0.000400
300 Maximum 0.166400
Minimum -0.132300
200 Std. Dev. 0.021331
Skewness 0.103542
100 Kurtosis 7.650982

0 Jarque-Bera 2231.572
-0.10 -0.05 0.00 0.05 0.10 0.15 Probability 0.000000
HETEROSCEDASTICITY
Heteroskedasticity Test: ARCH

F-statistic 15.00314 Prob. F(1,2465) 0.0001

Obs*R-squared 14.92448 Prob. Chi-Square(1) 0.0001

Test Equation:
Dependent Variable: RESID^2
Method: Least Squares
Date: 09/15/24 Time: 10:29
Sample (adjusted): 3 2472
Included observations: 2467 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

C 0.000419 2.52E-05 16.61184 0.0000

RESID^2(-1) 0.077777 0.020080 3.873389 0.0001

R-squared 0.006050 Mean dependent var 0.000455

Adjusted R-squared 0.005646 S.D. dependent var 0.001172
S.E. of regression 0.001169 Akaike info criterion -10.66537
Sum squared resid 0.003366 Schwarz criterion -10.66066
Log likelihood 13157.73 Hannan-Quinn criter. -10.66366
F-statistic 15.00314 Durbin-Watson stat 2.008110
Prob(F-statistic) 0.000110

MULTICOLLINERAITY

Variance Inflation Factors

Date: 09/15/24 Time: 10:06
Sample: 1 2473
Included observations: 2469

Coefficient Uncentered Centered

Variable Variance VIF VIF

LOG___CRUDE_OIL_ 0.000223 1.000257 1.000000

C 1.84E-07 1.000257 NA
ARCH
Dependent Variable: LOG_PRICE_CRUDE_OIL
Method: ML ARCH - Normal distribution (BFGS / Marquardt steps)
Date: 09/15/24 Time: 11:57
Sample (adjusted): 8/06/2014 7/31/2024
Included observations: 2468 after adjustments
Convergence achieved after 11 iterations
Coefficient covariance computed using outer product of gradients
Presample variance: backcast (parameter = 0.7)
GARCH = C(3) + C(4)*RESID(-1)^2

Variable Coefficient Std. Error z-Statistic Prob.

LOG_PRICE_PERSISTENT_SYSTEMS -0.051535 0.016691 -3.087520 0.0020

C -0.066707 0.028663 -2.327230 0.0200

Variance Equation

C 4.493902 0.081067 55.43436 0.0000

RESID(-1)^2 0.522035 0.017010 30.69072 0.0000

R-squared 0.001043 Mean dependent var -0.038794

Adjusted R-squared 0.000638 S.D. dependent var 2.875150
S.E. of regression 2.874233 Akaike info criterion 4.721666
Sum squared resid 20372.15 Schwarz criterion 4.731085
Log likelihood -5822.536 Hannan-Quinn criter. 4.725088
Durbin-Watson stat 1.966745

24

20

16

12

0
14 16 18 20 22 24

Conditional standard deviation

GARCH
Dependent Variable: LOG_PRICE_CRUDE_OIL
Method: ML ARCH - Normal distribution (BFGS / Marquardt steps)
Date: 09/15/24 Time: 11:58
Sample (adjusted): 8/06/2014 7/31/2024
Included observations: 2468 after adjustments
Convergence achieved after 22 iterations
Coefficient covariance computed using outer product of gradients
Presample variance: backcast (parameter = 0.7)
GARCH = C(3) + C(4)*RESID(-1)^2 + C(5)*GARCH(-1)

Variable Coefficient Std. Error z-Statistic Prob.

LOG_PRICE_PERSISTENT_SYSTEMS -0.030702 0.018811 -1.632107 0.1027

C -0.085484 0.040338 -2.119181 0.0341

Variance Equation

C 0.200655 0.028147 7.128885 0.0000

RESID(-1)^2 0.119980 0.007088 16.92704 0.0000
GARCH(-1) 0.851417 0.009249 92.05032 0.0000

R-squared 0.000783 Mean dependent var -0.038794

Adjusted R-squared 0.000377 S.D. dependent var 2.875150
S.E. of regression 2.874608 Akaike info criterion 4.476532
Sum squared resid 20377.47 Schwarz criterion 4.488305
Log likelihood -5519.041 Hannan-Quinn criter. 4.480809
Durbin-Watson stat 1.966794

TARCH
Dependent Variable: LOG_PRICE_CRUDE_OIL
Method: ML ARCH - Normal distribution (BFGS / Marquardt steps)
Date: 09/15/24 Time: 11:58
Sample (adjusted): 8/06/2014 7/31/2024
Included observations: 2468 after adjustments
Convergence achieved after 16 iterations
Coefficient covariance computed using outer product of gradients
Presample variance: backcast (parameter = 0.7)
GARCH = C(3) + C(4)*RESID(-1)^2 + C(5)*RESID(-1)^2*(RESID(-1)<0)

Variable Coefficient Std. Error z-Statistic Prob.

LOG_PRICE_PERSISTENT_SYSTEMS -0.054864 0.017446 -3.144814 0.0017

C -0.018306 0.043948 -0.416546 0.6770

Variance Equation

C 4.543701 0.090511 50.20077 0.0000

RESID(-1)^2 0.641142 0.030053 21.33385 0.0000
RESID(-1)^2*(RESID(-1)<0) -0.271475 0.043503 -6.240345 0.0000

R-squared 0.000988 Mean dependent var -0.038794

Adjusted R-squared 0.000583 S.D. dependent var 2.875150
S.E. of regression 2.874312 Akaike info criterion 4.718904
Sum squared resid 20373.28 Schwarz criterion 4.730677
Log likelihood -5818.127 Hannan-Quinn criter. 4.723181
Durbin-Watson stat 1.966631
EARCH
Dependent Variable: LOG_PRICE_CRUDE_OIL
Method: ML ARCH - Normal distribution (BFGS / Marquardt steps)
Date: 09/15/24 Time: 11:59
Sample (adjusted): 8/06/2014 7/31/2024
Included observations: 2468 after adjustments
Failure to improve likelihood (singular hessian) after 17 iterations
Coefficient covariance computed using outer product of gradients
Presample variance: backcast (parameter = 0.7)
LOG(GARCH) = C(3) + C(4)*ABS(RESID(-1)/@SQRT(GARCH(-1))) +
C(5)*RESID(-1)/@SQRT(GARCH(-1))

Variable Coefficient Std. Error z-Statistic Prob.

LOG_PRICE_PERSISTENT_SYSTEMS -0.057806 0.024091 -2.399472 0.0164

C -0.039956 0.047970 -0.832941 0.4049

Variance Equation

C(3) 1.658052 0.013924 119.0811 0.0000

C(4) 0.435573 0.013968 31.18438 0.0000
C(5) 0.035882 0.012432 2.886212 0.0039

R-squared 0.001034 Mean dependent var -0.038794

Adjusted R-squared 0.000629 S.D. dependent var 2.875150
S.E. of regression 2.874246 Akaike info criterion 4.795296
Sum squared resid 20372.34 Schwarz criterion 4.807069
Log likelihood -5912.396 Hannan-Quinn criter. 4.799573
Durbin-Watson stat 1.966736

28

24

20

16

12

0
14 16 18 20 22 24

Conditional standard deviation