search

bcgov / ssdtools

An R package to fit and plot Species Sensitivity Distributions (SSDs)
https://bcgov.github.io/ssdtools/
Apache License 2.0
30 stars 16 forks source link

More information about the ssd_gof #251

Closed liqian-Rcees closed 2 years ago

liqian-Rcees commented 2 years ago

Hi, I wonder whether any code to get the R Square and the mean standard error of the fitness of different distribution？ thank you！

joethorley commented 2 years ago

@liqian-Rcees R Square measures the variation in the data that is explained by one or more predictors. In the current context there are no predictors as the models simply describe the variation in the data so R2 is not applicable.

See also issue #111

I'm not sure what the mean standard error of the fitness is.

liqian-Rcees commented 2 years ago

@joethorley Thank you for your response, I wonder whether R2 can be reckoned by a pair of the collected toxicology data and the its responding predicted value at the same proportion of species affected in a certain distribution. mean standard error (MSE) is similar to RSS #111 . if these can be added to the gof function, results may provide more information and choices for users. This is just a suggestion, I also referred to the EPA SSDgenerator based on log-norm dist, thank you! https://view.officeapps.live.com/op/view.aspx?src=https%3A%2F%2Fwww.epa.gov%2Fsites%2Fdefault%2Ffiles%2F2017-10%2Fssd_generator_v1.xlsm&wdOrigin=BROWSELINK

joethorley commented 2 years ago

@liqian-Rcees - this is outside the scope of the current software