Consider the following ARMA model for {y,}: P Σ a +E + b j=1 1=1 where {e } is the residual. Which of the following assumption(s) on the residuals {ɛ } is needed to enable forecasting with this model...

Extracted text: Consider the following ARMA model for {y,}: P Σ a +E + b j=1 1=1 where {e } is the residual. Which of the following assumption(s) on the residuals {ɛ } is needed to enable forecasting with this model in practice? O a. e is normally distributed. Ob. e is mean-independent of y,1,-2"* and e 1-1' C. and e are stochastically independent for all +s: O d. All of the above. QUESTION 2 How can the validity of the assumption(s) identified in Question 1 be examined in practice? O a. Use the Breusch-Pagan test, where the null hypothesis is that the distribution is normal. b. Use the Ljung-Box test, where the null hypothesis is that the first K autocorrelations are zero. Examine the SACF and SPACF of the estimated residuals d. Both (b) and (c).