Fig. 2
From: Multiple imputation and direct estimation for qPCR data with non-detects
Comparison of methods for non-detects based on masked at Cq=30 values. The left panel shows the difference between \(\theta\) estimated from the complete data and the estimate of \(\theta\) obtained from the truncated data, applying PEMM, DirEst, and MI respectively. The right panel shows the difference between \(\sigma ^2\) calculated from the complete data and four estimates of \(\sigma ^2\): from the truncated data, PEMM, DirEst, and MI. Black solid lines represent means of the differences between parameter estimates from the complete data and parameter estimates from the masked data