This paper develops generalized method of moments-based (GMM-based) Lagrange multiplier tests for nonlinear hypotheses that are robust to locally misspecified possibly nonlinear alternatives. The procedure is based on an initial consistent GMM estimator of the parameters under a given set of nonlinear restrictions. The new test for one particular set of nonlinear hypotheses is consistent and has correct asymptotic size independently of whether the other, also nonlinear hypotheses, are correct or locally misspecified. To illustrate the usefulness of our proposed tests we consider testing rational expectations hypotheses using U.S. data.

中文翻译：

---

测试错误指定条件下的非线性限制条件，并用于检验理性预期假设

本文针对非线性假设开发了基于矩（基于GMM）的Lagrange乘数检验的通用方法，该方法对于局部错误指定的可能非线性选择具有鲁棒性。该程序基于给定的非线性限制条件下参数的初始一致GMM估计量。针对一组特定的非线性假设的新检验是一致的，并且具有正确的渐近大小，而与另一个以及非线性假设是否正确或局部错误指定无关。为了说明我们提出的检验的有用性，我们考虑使用美国数据检验理性预期假设。