gretl version 2017d-git Current session: 2017-09-08 22:39 ? open phillips_aus.gdt Read datafile C:\Users\x\AppData\Roaming\gretl\data\POE4\phillips_aus.gdt periodicity: 4, maxobs: 91 observations range: 1987:1 to 2009:3 Listing 3 variables: 0) const 1) inf 2) u ? eval $sysinfo bundle anonymous: nproc = 2 blascore = Nehalem hostname = HER-PC os = windows mpi = 0 blas = openblas omp_num_threads = 2 omp = 1 blas_parallel = OpenMP mpimax = 2 wordlen = 64 ? arma 1 1; inf 0 diff(u) --verbose ARMA initialization: using nonlinear AR model Iteration 1: loglikelihood = -67.4802393457 Parameters: 0.76087 0.55739 0.00010000 -0.69439 Gradients: 1.7670 0.13716 -11.827 -0.17685 (norm 6.21e-001) MA root 0 = 0.00714939 MA estimate(s) out of bounds MA root 0 = 0.178747 MA estimate(s) out of bounds Iteration 2: loglikelihood = -66.8018116428 (steplength = 0.008) Parameters: 0.77501 0.55849 -0.094515 -0.69580 Gradients: 0.86167 11.492 -1.7851 0.030193 (norm 1.35e+000) MA root 0 = 0.004081 MA estimate(s) out of bounds MA root 0 = 0.0972698 MA estimate(s) out of bounds Iteration 3: loglikelihood = -65.2733116526 (steplength = 0.008) Parameters: 0.79933 0.67040 -0.21899 -0.69712 Gradients: -0.59560 10.317 -3.0064 -0.46834 (norm 1.45e+000) MA root 0 = 1.56421e-007 MA estimate(s) out of bounds MA root 0 = 3.90781e-006 MA estimate(s) out of bounds MA root 0 = 9.73578e-005 MA estimate(s) out of bounds MA root 0 = 0.00239241 MA estimate(s) out of bounds MA root 0 = 0.0549967 MA estimate(s) out of bounds MA root 0 = 0.946232 MA estimate(s) out of bounds Iteration 4: loglikelihood = -63.4855287291 (steplength = 6.4e-005) Parameters: 0.82010 0.81500 -0.38079 -0.70208 Gradients: -0.46740 1.9059 -6.9354 -1.1323 (norm 1.16e+000) MA root 0 = 0.670997 MA estimate(s) out of bounds Iteration 5: loglikelihood = -63.2029863976 (steplength = 0.04) Parameters: 0.84885 0.84329 -0.41439 -0.77209 Gradients: -1.0790 -0.33677 -7.4538 -0.00059934 (norm 1.04e+000) MA root 0 = 0.622613 MA estimate(s) out of bounds Iteration 6: loglikelihood = -62.9895429927 (steplength = 0.04) Parameters: 0.85054 0.86582 -0.44851 -0.74957 Gradients: -0.76270 -2.7956 -7.4380 -0.46318 (norm 1.30e+000) Iteration 7: loglikelihood = -62.8310629234 (steplength = 1) Parameters: 0.91930 0.94758 -0.61982 -0.75593 Gradients: -0.32522 -14.305 -0.061113 -0.13515 (norm 1.87e+000) Iteration 8: loglikelihood = -62.6476395652 (steplength = 1) Parameters: 0.89467 0.89226 -0.54545 -0.73749 Gradients: -1.6020 4.9033 0.076397 -0.48887 (norm 1.25e+000) Iteration 9: loglikelihood = -62.6002115983 (steplength = 1) Parameters: 0.88668 0.90381 -0.55164 -0.74886 Gradients: -1.0203 -0.082869 -1.4221 -0.33966 (norm 7.10e-001) Iteration 10: loglikelihood = -62.5854811880 (steplength = 1) Parameters: 0.88515 0.91328 -0.57091 -0.75282 Gradients: -0.76536 -1.4956 -0.57813 -0.24313 (norm 7.99e-001) MA root 0 = 0.757417 MA estimate(s) out of bounds Iteration 11: loglikelihood = -62.5824367401 (steplength = 0.0016) Parameters: 0.88392 0.91089 -0.57183 -0.75321 Gradients: -0.82415 0.044644 0.046904 -0.21440 (norm 4.89e-001) Iteration 12: loglikelihood = -62.5631679322 (steplength = 0.04) Parameters: 0.84155 0.90914 -0.57105 -0.76435 Gradients: -0.0079279 -0.11055 0.23221 -0.0013639 (norm 2.45e-001) Iteration 13: loglikelihood = -62.5627574065 (steplength = 0.008) Parameters: 0.84098 0.90833 -0.56911 -0.76449 Gradients: -0.0082188 -0.070007 0.10926 -0.0026628 (norm 1.84e-001) Iteration 14: loglikelihood = -62.5626197483 (steplength = 1) Parameters: 0.84010 0.90746 -0.56714 -0.76558 Gradients: -0.0029615 -0.0029232 0.00073825 0.014548 (norm 6.46e-002) Iteration 15: loglikelihood = -62.5626145035 (steplength = 0.04) Parameters: 0.83993 0.90742 -0.56709 -0.76505 Gradients: 0.00015028 0.0012260 -0.0014360 0.0044334 (norm 3.69e-002) Iteration 16: loglikelihood = -62.5626139473 (steplength = 1) Parameters: 0.83996 0.90744 -0.56713 -0.76482 Gradients: -6.8567e-005 -0.00034497 -0.00021174-2.5580e-005 (norm 1.13e-002) Iteration 17: loglikelihood = -62.5626139470 (steplength = 1) Parameters: 0.83995 0.90744 -0.56713 -0.76482 Gradients: 8.5265e-006 9.9476e-006 1.2434e-005-7.1054e-007 (norm 2.44e-003) Iteration 17: loglikelihood = -62.5626139470 (steplength = 1) Parameters: 0.83995 0.90744 -0.56713 -0.76482 Gradients: 8.5265e-006 9.9476e-006 1.2434e-005-7.1054e-007 (norm 2.44e-003) --- FINAL VALUES: loglikelihood = -62.5626139470 (steplength = 0.00032) Parameters: 0.83995 0.90744 -0.56713 -0.76482 Gradients: 8.5265e-006 9.9476e-006 1.2434e-005-7.1054e-007 (norm 2.44e-003) Function evaluations: 51 Evaluations of gradient: 17 Model 1: ARMAX, using observations 1987:2-2009:3 (T = 90) Estimated using Kalman filter (exact ML) Dependent variable: inf Standard errors based on Hessian coefficient std. error z p-value -------------------------------------------------------- const 0.839953 0.223287 3.762 0.0002 *** phi_1 0.907436 0.0628997 14.43 3.51e-047 *** theta_1 −0.567130 0.119891 −4.730 2.24e-06 *** d_u −0.764820 0.229142 −3.338 0.0008 *** Mean dependent var 0.791111 S.D. dependent var 0.636819 Mean of innovations −0.018008 S.D. of innovations 0.483088 Log-likelihood −62.56261 Akaike criterion 135.1252 Schwarz criterion 147.6243 Hannan-Quinn 140.1656 Real Imaginary Modulus Frequency ----------------------------------------------------------- AR Root 1 1.1020 0.0000 1.1020 0.0000 MA Root 1 1.7633 0.0000 1.7633 0.0000 -----------------------------------------------------------