In expression (5) we have no information that the variance of the sum is the sum of the
variances since all the innovations are time t innovations, which may be correlated. In fact, for given a6, a, the maximum variance of the sum in (5) occurs when the elements in
the sum are perfectly positively correlated.This means then that so long as var(Sd)#O0,
td,+k =akS(d(, where ak =Gk/GO- Substituting this into (6) implies