Hildreth-Lu Search Procedure:
Since we know that p lies between 1 and - 1, this procedure searches over this range, i.e., using values of p say between -0.9 and 0.9 in intervals of 0.1. For each ρ we compute the regression in (7.3) and report the residual sum of squares corresponding to that ρ. The minimum residual sum of squares gives us our choice of ρ and the corresponding regression gives us the estimates of β1,β2 and σ2. We can refine this procedure around the best ρ found in the first stage of the search. For example, suppose that ρ=0.6 gave the-minimum residual sum of squares. We can search next between 0.5 1 and 0.69 in intervals of 0.01. This search procedure guards against a local minimum.