The smoothing equation is Lt= ˛yt+ (1 − ˛)Lt−1, where Ltis the forecast value of yield for next period at time t; Lt−1is the actual yield at time t − 1; ˛ is the smoothing constant (0 < ˛ < 1). Values of ˛close to one have less of a smoothing effect and give greater weight to recent changes in the data, while values of ˛ closer to zero have agreater smoothing effect and are less responsive to recent changes.For each series of yields the value of ˛ was used, which gave the model with a minimum mean square error