[Gretl-devel] a very strange case of arima (poor)convergence

Dear all, The script below illustrate the problem Findings: extremely bad lnl in comparison to --x-12-arima zero values of the 2 last parameters and gradients at all iterations Strangely large scaling factor Note that --x-12-arima gives nice pol. roots and excellent Ljung-Box Q' Also note that with the default $pd for modtest --autocorr obviously fails because zero df open bad_data.gdt #attached smpl 1 194 # note strange zeros for b[y_one] and b[y_two] arima 3 0 0; 1 0 0; diff_series const y_one y_two   lnl1 = $lnl modtest --autocorr 5 # compare arima 3 0 0; 1 0 0; diff_series const y_one y_two --x-12-arima lnl2 = $lnl modtest --autocorr 5 eval lnl2 - lnl1 arima 3 0 0; 1 0 0; diff_series const y_one y_two --verbose # Scaling y by 2.18989e+018 !!! /* Iteration 1: loglikelihood = -8292.74884180 Parameters: -2.8152e+016     0.65496   -0.090036    -0.16211    -0.44328     0.00000                 0.00000 Gradients:   4.7608e-018      4.0881      1.2538      3.3817     -1.2644     0.00000                 0.00000 (norm 7.59e-001) Iteration 2: loglikelihood = -8292.71644305 (steplength = 0.0016) Parameters: -2.8152e+016     0.66150   -0.088030    -0.15670    -0.44531     0.00000                 0.00000 Gradients:   4.5624e-018      1.3507     -1.5003      1.0847     -1.4772     0.00000                 0.00000 (norm 5.32e-001) */ #etc. etc Oleh