Consider the model Yt = α + β1 X1t + β2 X2t + β3 Yt-1 + vt
Suppose Yt-1 and vt are correlated. To remove the correlation, suppose we use the following instrumental variable approach: First regress Yt on X1t and X2t and obtain the estimated Yˆt from this regression. Then regress Yt = α + β1 X1t + β2 X2t + β3 Yˆt-1 + vt
where Yˆt-1 are estimated from the ?rst-stage regression.
a. How does this procedure remove the correlation between Yt-1 and vt in the original model?
b. What are the advantages of the recommended procedure over the Liviatan approach?