Use INTQRT.RAW for this exercise.
(i) In Example 18.7, we estimated an error correction model for the holding yield on six-month T-bills, where one lag of the holding yield on three-month T-bills is the explanatory variable. We assumed that the co integration parameter was one in the equation hy6t = α + βhy3t-1 + ut. Now, add the lead change, Δhy3t, the contemporaneous change, Δhy3t-1, and the lagged change, Δhy3t-2, of hy3t-1. That is, estimate the equation
hy6t = α + βhy3t-1 + ϕ0 Δhy3t + ϕ1 Δhy3t-1 + pl Δhy3t-2 + et
and report the results in equation form. Test HQ: f3 = 1 against a two-sided alternative. Assume that the lead and lag are sufficient so that {hy3t-1} is strictly exogenous in this equation and do not worry about serial correlation.
(ii) To the error correction model in (18.39), add Δ/ry3t-2 and (hy6t-2 - hy3t-3). Are these terms jointly significant? What do you conclude about the appropriate error correction model?