Table 3. Spearman's correlation of CANO estimates across models for the CHS data.
Model | linP >30 | linT >30 | quadP | quadT | nonLin | nonLinLog | nonLinLogC | HMA | Condorelli | Kerckx |
linP >30 | 1 | |||||||||
linT >30 | 0.91 | 1 | ||||||||
quadP | 0.75 | 0.71 | 1 | |||||||
quadT | 0.63 | 0.82 | 0.75 | 1 | ||||||
nonLin | 0.67 | 0.65 | 0.98 | 0.75 | 1 | |||||
nonLinLog | 0.57 | 0.74 | 0.77 | 0.98 | 0.78 | 1 | ||||
nonLinLogC | 0.57 | 0.75 | 0.77 | 0.98 | 0.79 | 1.00 | 1 | |||
HMA | 0.48 | 0.74 | 0.54 | 0.94 | 0.54 | 0.90 | 0.91 | 1 | ||
Condorelli | −0.01 | 0.25 | 0.25 | 0.58 | 0.29 | 0.60 | 0.60 | 0.67 | 1 | |
Kerckx | 0.55 | 0.53 | 0.73 | 0.56 | 0.71 | 0.57 | 0.58 | 0.42 | 0.48 | 1 |