QUESTION FOUR (a) Define covariance stationarity and explain how the correlogram can be constructed. (b) Express the equations describing AR(1), MA(1), and ARMA(1,1) processes and explain when such processes are stationary. Outline the characteristics of the autocorrelation function for a white noise process. (c) Suppose that you estimate an AR(2) model and obtain the following results: yt = 0.039 + 0.77yt-1 + 0.09yt-2 + ut. Assuming that yT-1 = 4.27 and yT-2 = 3.52, produce the forecasts of y for T and T +1. (d) Explain carefully how to evaluate forecasting performance, in general.