Page 9 - Azerbaijan State University of Economics
P. 9

Asif  Alıyev: Forecasting of Engel Curve Components with the Application of ARIMA Method

                    Table 2: ARIMA model for household incomes
                    Dependent Variable: D(INCOME)
                    Method: Least Squares
                    Date: 01/08/23   Time: 10:57
                    Sample (adjusted): 3 14
                    Included observations: 12 after adjustments
                    Convergence achieved after 17 iterations
                    MA Backcast: 2
                    Variable                  Coefficient    Std. Error     t-Statistic    Prob.



                      C                        3.475553       13.10608      0.265186      0.7968

                    AR(1)                      0.879029       0.128212      6.856054      0.0001
                    MA(1)                      -0.999931      0.155679      -6.423015     0.0001
                      R-squared                0.589464  Mean dependent var              14.63333



                    Adjusted R-squared         0.498234  S.D. dependent var              7.604464
                    S.E. of regression         5.386656  Akaike info criterion           6.418045
                    Sum squared resid          261.1446  Schwarz criterion               6.539271
                    Log likelihood             -35.50827  Hannan-Quinn criter.           6.373162
                    F-statistic                6.461285  Durbin-Watson stat              2.438545
                    Prob(F-statistic)          0.018200

                    Table 3: ARIMA model for household expenses
                    Dependent Variable: D(EXPENSES)
                    Method: Least Squares
                    Date: 01/08/23   Time: 12:58
                    Sample (adjusted): 7 14
                    Included observations: 8 after adjustments
                    Failure to improve SSR after 13 iterations
                    MA Backcast: 5 6
                    Variable                  Coefficient    Std. Error     t-Statistic     Prob.
                    C                          -1.146122      21.55672      -0.053168      0.9609
                    AR(1)                      0.095612       0.419764      0.227775       0.8345
                    AR(5)                      0.489263       0.653582      0.748587       0.5085
                    MA(1)                      -0.646581      1.235461      -0.523352      0.6369
                    MA(2)                      0.999303       1.159314      0.861978       0.4521
                    R-squared                  0.783570  Mean dependent var               10.90000
                    Adjusted R-squared         0.494997  S.D. dependent var               5.668459
                    S.E. of regression         4.028208  Akaike info criterion            5.893691
                    Sum squared resid          48.67938  Schwarz criterion                5.943342
                    Log likelihood             -18.57476  Hannan-Quinn criter.            5.558815
                    F-statistic                2.715327  Durbin-Watson stat               2.341917
                    Prob(F-statistic)          0.219031



                                                            9
   4   5   6   7   8   9   10   11   12   13   14