Comments on "Gradient estimation algorithms for equation error formulations"

It is shown that the optimal value for the gain of a gradient parameter estimator derived in the above paper is, in general, not optimal, and for a simple example the actual optimal gain is determined.