Confidence interval for random-effects calibration curves with left-censored data

The presence of left censoring in environmental and engineering applications complicates tests of hypotheses and interval estimation. First, we construct a confidence interval (CI) for a true concentration using nonlinear mixed-effects regression models for analysis of interlaboratory calibration data with left censoring. Next, we impute the unobserved left-censored data with randomly simulated value. Finally, we compute the coverage probability via simulation and show that our methodology achieves the nominal level. We illustrate our methodology with a real-life example.