Table 1 Optimum parameter values for the imputation methods in our study. BPCA and LLS did not require parameter optimization.
Method | Parameter/Values tested | Optimal Value | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
ALI | ALO | BAL | CAU | GAS | GOL | ROS | SP.AFA | SP.ELU | ||
KNN.e | K = (5,10,15,25,50,100,200) | 5 | 5 | 10 | 15 | 10 | 15 | 5 | 10 | 15 |
KNN.c | K = (5,10,15,25,50,100,200) | 10 | 10 | 15 | 15 | 10 | 15 | 5 | 15 | 25 |
OLS.c | K = (5,10,15,25,50,100,200) | 10 | 10 | 10 | 15 | 5 | 10 | 5 | 25 | 200 |
PLS | C = (2,3,4,5,6,7,8,9,10,15,25) | 8 | 10 | 15 | 8 | 10 | 5 | 15 | 8 | 9 |
SVD | p = 0.05 to 0.5, by 0.05 | 0.1 | 0.25 | 0.2 | 0.25 | 0.15 | 0.15 | 0.15 | 0.2 | 0.15 |
LLS | k (values tested depended on data set) | 710 | 710 | 710 | 2397 | 2397 | 1598 | 710 | 210 | 210 |