Skip to main content
. Author manuscript; available in PMC: 2020 Nov 5.
Published in final edited form as: Environ Sci Technol. 2019 Oct 10;53(21):12793–12802. doi: 10.1021/acs.est.9b03957

Table 5:

External validation benchmark against ECOSAR.

Model Used Test Set Size Train Set Size R2 RMSEi σE lower σEii σE upper Percent |Error| < 1
LC50 ECOSAR 532 0.23 1.07 1.01 1.07 1.14 83.6%
Fast Model 532 2124 0.61 0.76 0.72 0.76 0.81 84.8%
Chronic MATC ECOSAR 18 0.69 0.62 0.46 0.61 0.91 94.4%
Fast Model 18 1908 0.78 0.52 0.40 0.54 0.80 94.4%
Expanded chronic MATC ECOSAR 45 −0.60 1.50 1.24 1.50 1.89 66.7%
Fast Model 45 1881 0.43 0.89 0.75 0.90 1.14 80.0%
i

RMSE: Root-mean-square error

ii

σE: Standard deviation of error