Based on the AIC criteria, we select the ARMA(2,0) model using the detrended time series shown above. The second order polynomial model is given by Yt = 11.22 - 177.82t + 265t2 + Xt, where Xt is an ARMA(2,0). Table 1 shows the estimated ARMA model: Mean Estimate 0.9541 -0.3074 -0.0111 s.e. 0.0975 0.0979 0.1865 Table 1: Estimated ARMA Model When performing t-tests to assess the significance of the estimated coefficients, what degrees of freedom would you use to determine the appropriate cutoff value for a .95 level of confidence. Note: Note that the calculation of the degrees of freedom in this example follows from the linear regression modeling, however, this is not quite correct since the regression analysis model assumes independence among the observations. An important follow up questions are: where do we need the computation of the degrees of freedom and how will the incorrect degrees of freedom affect it? O 1.96 O 95 O 96 O 98