Consider the following model: yt = 0 + 1yt1 + ut ; ut = ut1 + et where et i.i.d(0, 2 e ).

The regression output using a sample of 1200 observations is given by: yt = 0.9 + 0.7yt1; ut= 0.2utt1 + et with var d(b) = 0.0064.

i) Based on the regression output provided above, test whether the model suffers from AR(1) serial correlation at the 5% significance level using TWO different methods. Do both methods/tests give the same conclusion regarding the presence of an AR(1) serial correlation? For each testing method, state the null hypothesis, compute the test, and compare the test statistic to the critical value(s).

ii) Assume that an economist suggested that the model suffers from AR(2) serial correlation, and you were asked to assess his claim using the information provided below. utt = 0.12ut1 + 0.37ubt2; R 2 ut = 0.0045. Do you agree with the economist? Show your work(state the null, compute the test statistic, and compare it with the relevant critical value(s)).