This paper studies the estimation and inference of nonlinear econometric models when the economic variables are contained in different data sets. We construct a semiparametric minimum distance (SMD) estimator of the unknown structural parameter of interest when there are some common conditioning variables in different data sets. The SMD estimator is shown to be consistent and has an asymptotic normal distribution. We provide the explicit form of the optimal weight for the SMD estimation. We provide a consistent estimator of the variance–covariance matrix of the SMD estimator, and hence inference procedures of the unknown parameter vector. The finite sample performances of the SMD estimators and the proposed inference procedures are investigated in few alternative Monte Carlo simulation studies.

本文研究了当经济变量包含在不同的数据集中时非线性计量经济模型的估计和推理。当不同数据集中存在一些共同的条件变量时，我们构建了感兴趣的未知结构参数的半参数最小距离 (SMD) 估计量。SMD 估计量被证明是一致的并且具有渐近正态分布。我们为 SMD 估计提供了最佳权重的显式形式。我们提供了 SMD 估计量的方差-协方差矩阵的一致估计量，从而提供了未知参数向量的推理过程。SMD 估计器的有限样本性能和提议的推理程序在少数替代蒙特卡罗模拟研究中进行了研究。