Explain based on (i) whether the models are covariance stationary and invertible, (ii) if your residuals are white noise, and (iii) based on the AIC and SIC which of the three models above should be preferred. Justify your answer.

Dependent Variable: GC Method: ARMA Maximum Likelihood (OPG - BHHH) Date: 11/01/22 Time: 20:54 Sample: 19802017 Included observations: 38 Convergence achieved after 17 iterations Coefficient covariance computed using outer product of gradients d.f. adiustment for standard errors \& covariance Dependent Variable: GC Method: ARMA Maximum Likelihood (OPG - BHHH) Date: 11/01/22 Time: 21:03 Sample: 19802017 Included observations: 38 Convergence achieved after 28 iterations Coefficient covariance computed using outer product of gradients df adiusitment for standard errors \& covariance. Dependent Variable: GC Method: ARMA Maximum Likelihood (OPG - BHHH) Date: 11/01/22 Time: 21:07 Sample: 19802017 Included observations: 38 Convergence achieved after 21 iterations Coefficient covariance computed using outer product of gradients Dependent Variable: GC Method: ARMA Maximum Likelihood (OPG - BHHH) Date: 11/01/22 Time: 20:54 Sample: 19802017 Included observations: 38 Convergence achieved after 17 iterations Coefficient covariance computed using outer product of gradients d.f. adiustment for standard errors \& covariance Dependent Variable: GC Method: ARMA Maximum Likelihood (OPG - BHHH) Date: 11/01/22 Time: 21:03 Sample: 19802017 Included observations: 38 Convergence achieved after 28 iterations Coefficient covariance computed using outer product of gradients df adiusitment for standard errors \& covariance. Dependent Variable: GC Method: ARMA Maximum Likelihood (OPG - BHHH) Date: 11/01/22 Time: 21:07 Sample: 19802017 Included observations: 38 Convergence achieved after 21 iterations Coefficient covariance computed using outer product of gradients