Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Since the actual solution to intertemporal rational expectations models is usually not known, it is useful to have criteria for judging the accuracy of a given numerical solution. In this paper we propose a test for accuracy that is easy to implement and can be applied to a wide class of models without knowledge of the exact solution. We discuss the power of the test by simulating several models with the linear-quadratic approximation and with the method of parameterized expectations. We conclude that the test is powerful.