Given below are Microsoft Excel outputs for various estimated autoregressive (AR) models for Australia's greenhouse gas - CO2 emissions (metric tons per capita) from 2000 to 2020. From the data, you also know that the real CO2 emissions (metric tons per capita) in Australia for 2018, 2019 and 2020 are 15.49, 15.24 and 15.22, respectively. AR(1) Model: Coefficients Standard Error t Stat P-value Intercept -0.739254405 1.058826038 -0.69818 0.49352 Yi-1 1.03633313 0.061498926 16.85124 0.00000 AR(2) Model: Coefficients Standard Error t Stat P-value Intercept -0.195598275 1.195728319 -0.16358 0.8719 Yi-1 1.254042159 0.230101005 5.449964 0.0000 Yi-2 -0.247748864 0.252306745 -0.98194 0.3391 AR(3) Model: Coefficients Standard Error t Stat P-value Intercept 0.267661461 1.27502274 0.209927 0.83622 Yi-1 1.193189828 0.237113007 5.032157 0.000102 Yi-2 -0.006585142 0.343435519 -0.01917 0.984925 Yi-3 -0.206841239 0.200271827 -1.0328 0.316166 Using a 5% level of significance, what is the appropriate autoregressive model (AR model) for the CO2 emissions (metric tons per capita) in Australia? Explain your reasoning. If you decide to use the second-order autoregressive model (AR(2) Model), what will the predicted Australia's CO2 emissions (metric tons per capita) be in 2020? Show your calculations