Question 3 Given two time series for the year 2020: Month January February March April May June -1.049 -1.075 0.309 0.679 1.233 0.560 0 0 1 1 Compute the forecasted values for Y, for July and August in 2020 by using the models stated in (a) to (d): (a) Moving average of order 4. (b) Single exponential smoothing with discount factor 0.2 (Remark: Use the January value as initial value). (c) The following ARIMA(0,0,1) model: Y= = -0.7 + &+ + 0.581-1 (d) The following dynamic regression: 0.7 Y. = 1 - 0.2B 54t-2 - 0.5Y-1 + Et (e) Finally, given the observed values of July and August 2020: Month July August It -0.587 0.191 Xt 1 1 Evaluate the forecasting accuracy of the models used in (a), (b), (c) and (d) by their sum of squared deviation. Here, the sum of squared deviation only needs to be calculated based on the forecasted values for July and August