We investigate the performance of a recently developed algorithm that evaluates the uncertainty of nonlinear multivariate microwave calibration models using regression residuals. We apply the algorithm to synthetic data consisting of both random and systematic errors and show that the algorithm can account for both types of errors even in the absence of accurate models for the random errors. We also verify the algorithm with measured data.

中文翻译：

---

用回归残差评估非线性微波校准模型的不确定性


我们研究了最近开发的算法的性能，该算法使用回归残差评估非线性多元微波校准模型的不确定性。我们将该算法应用于由随机误差和系统误差组成的合成数据，并表明即使在缺乏随机误差的准确模型的情况下，该算法也可以解释这两种类型的误差。我们还用测量数据验证了算法。