The effect which the speed of adjustment parameter has on the statisti
cal properties of the partial adjustment model, and estimates of its p
arameters, is investigated. It is shown that in the case of very rapid
adjustment, the model approaches a classical (static) regression mode
l, but in the case of very slow adjustment, the dependent (or state) v
ariable displays near random walk behaviour. The finite sample perform
ance of the nonlinear least squares estimator is investigated in a sim
ulation study, and it is found that substantial bias and mean squared
error can be a feature of the estimates in the model with very slow ad
justment. This suggests that extreme caution should be exercised in si
tuations where the estimated speed of adjustment parameter is found to
be small.