Based on the

output, which model would you suggest?

Model I

Model II

Model III

Model IV

Model I ACF of Residuals Model III ACF of Residuals Coefficients: Coefficients: 0.3 ar1 ma ar1 0.923 0. 503 0. 952 0.1 s. e. 0.037 0. 070 s. e. 0. 028 -0.2 0.0 0.2 0.4 0.6 ACF ACF sigma*2 estimated as 1.07 sigma 2 estimated as 1.5 log likelihood = -146.8, aic = 297.7 log likelihood = -163.5, aic = 329 -0.3 10 15 20 10 15 20 Lag Lag Model II ACF of Residuals Model IV ACF of Residuals Coefficients: Coefficients: ar1 ar2 ari ar2 ar3 mal ma2 0.1 0.2 1.549 -0. 626 1.517 -0. 354 -0.240 -0.097 -0.238 s. e. 0. 076 0. 076 s. e. 0. 428 0.750 0. 355 0. 415 0. 174 ACE ACF -0.2 -0.1 0.0 0.1 0.2 sigma 2 estimated as 0.902 sigma*2 estimated as 0.846 -0.2 -0.1 0.0 log likelihood = -138.4, aic = 280.8 log likelihood = -135.3, aic = 280.7 5 10 15 20 5 10 15 20 Lag Lag