Business Forecasting Assignment

Name: Mansi Sharma

SAP Id: 80011220277
Roll No: A016
Group: 10
Step 1: Unit Root Test

The given data is not stationary. Thus, differencing is to be done to get a stationary data.
Null Hypothesis: PCE has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=11)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -0.207360 0.9325

Test critical values: 1% level -3.507394
5% level -2.895109
10% level -2.584738

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

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(PCE)
Method: Least Squares
Date: 07/01/21 Time: 16:11
Sample (adjusted): 1970Q2 1991Q4
Included observations: 87 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

PCE(-1) -0.000896 0.004322 -0.207360 0.8362

C 19.16937 11.10397 1.726352 0.0879

R-squared 0.000506 Mean dependent var 16.90345

Adjusted R-squared -0.011253 S.D. dependent var 18.29021
S.E. of regression 18.39283 Akaike info criterion 8.684519
Sum squared resid 28755.18 Schwarz criterion 8.741207
Log likelihood -375.7766 Hannan-Quinn criter. 8.707346
F-statistic 0.042998 Durbin-Watson stat 1.629810
Prob(F-statistic) 0.836225

The data is now stationary.

Null Hypothesis: D(PCE) has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=11)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -7.615082 0.0000

Test critical values: 1% level -3.508326
5% level -2.895512
10% level -2.584952

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

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(PCE,2)
Method: Least Squares
Date: 07/01/21 Time: 16:15
Sample (adjusted): 1970Q3 1991Q4
Included observations: 86 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(PCE(-1)) -0.820193 0.107706 -7.615082 0.0000

C 13.94370 2.689600 5.184300 0.0000

R-squared 0.408407 Mean dependent var -0.082558

Adjusted R-squared 0.401364 S.D. dependent var 23.49141
S.E. of regression 18.17567 Akaike info criterion 8.661026
Sum squared resid 27749.83 Schwarz criterion 8.718104
Log likelihood -370.4241 Hannan-Quinn criter. 8.683997
F-statistic 57.98948 Durbin-Watson stat 2.046675
Prob(F-statistic) 0.000000

On running the correlogram for the dataset at level, we got a gradual decay as shown below.

ACF – Gradual Decay

PACF – Significant Spike

Correlogram
Date: 07/01/21 Time: 16:17
Sample: 1970Q1 1991Q4
Included observations: 88
Autocorrelation Partial Correlation AC PAC Q-Stat Prob

1 0.970 0.970 85.581 0.000

2 0.938 -0.038 166.58 0.000
3 0.906 -0.019 243.04 0.000
4 0.872 -0.041 314.81 0.000
5 0.838 -0.030 381.87 0.000
6 0.802 -0.051 444.01 0.000
7 0.766 -0.019 501.36 0.000
8 0.729 -0.023 554.04 0.000
9 0.694 -0.010 602.28 0.000
10 0.658 -0.018 646.23 0.000
11 0.623 -0.002 686.21 0.000
12 0.590 -0.007 722.45 0.000
13 0.556 -0.021 755.10 0.000
14 0.523 -0.022 784.33 0.000
15 0.490 -0.014 810.33 0.000
16 0.456 -0.033 833.20 0.000
17 0.423 -0.018 853.13 0.000
18 0.389 -0.034 870.24 0.000
19 0.355 -0.018 884.73 0.000
20 0.321 -0.036 896.74 0.000
21 0.286 -0.035 906.44 0.000
22 0.253 -0.009 914.10 0.000
23 0.220 0.002 920.02 0.000
24 0.189 -0.010 924.46 0.000

The data is not stationary.

Unit root test at 1st difference.

AC: Gradual Decay

PAC: Significant Spike

Lag 3 is to be considered
Date: 07/01/21 Time: 16:19
Sample (adjusted): 1970Q2 1991Q4
Included observations: 87 after adjustments
Autocorrelation Partial Correlation AC PAC Q-Stat Prob

1 0.178 0.178 2.8542 0.091

2 0.191 0.165 6.1786 0.046
3 0.271 0.226 12.933 0.005
4 0.112 0.018 14.099 0.007
5 0.008 -0.095 14.105 0.015
6 0.065 -0.006 14.505 0.024
7 -0.035 -0.067 14.627 0.041
8 -0.135 -0.129 16.421 0.037
9 -0.070 -0.038 16.907 0.050
10 -0.046 0.030 17.117 0.072
11 -0.088 0.008 17.898 0.084
12 -0.065 -0.012 18.329 0.106
13 -0.032 -0.005 18.437 0.142
14 -0.230 -0.221 24.064 0.045
15 -0.093 -0.044 24.995 0.050
16 -0.023 0.045 25.052 0.069
17 -0.193 -0.105 29.166 0.033
18 -0.045 0.044 29.395 0.044
19 -0.117 -0.108 30.953 0.041
20 -0.081 0.008 31.710 0.046
21 -0.024 0.020 31.779 0.062
22 0.102 0.111 33.014 0.062
23 -0.108 -0.157 34.431 0.059
24 0.017 -0.007 34.467 0.077

Particulars ARIMA (1,1,1) ARIMA (1,1,3) ARIMA (3,1,3) ARIMA (3,1,1)

R Square 7.47% 8.37% 7.55% 8.59%
Adjusted R Square 4.13% 5.05% 4.20% 5.291%
No. of Significant 3 1 1 2
Coefficients
Volatility (Sigma SQ) 305.96 303.00 305.71 302.26
f-stat (corresponding p Not Significant Not Significant Not Significant Not Significant
value) (0.09) (0.06) (0.08) (0.057)
AIC 8.65 8.64 8.65 8.64
SIC 8.76 8.75 8.76 8.75
HQIC 8.70 8.69 8.7 8.68

ARIMA (1,1,1)
3,600

3,200

2,800

2,400

2,000

1,600
70 72 74 76 78 80 82 84 86 88 90

PCE PCEF