Confidence bands and intervals are valuable tools in evaluating how good a model fits data. Here's some reading for the technical/mathematical implementation. In addition, we need to think about the API. I suggest adding a method to either FitResults, or to Model.
Confidence bands and intervals are valuable tools in evaluating how good a model fits data. Here's some reading for the technical/mathematical implementation. In addition, we need to think about the API. I suggest adding a method to either
FitResults, or toModel.https://www.graphpad.com/support/faq/how-does-prism-compute-confidence-and-prediction-bands-for-nonlinear-regression/ https://stats.stackexchange.com/questions/85448/shape-of-confidence-and-prediction-intervals-for-nonlinear-regression https://stats.stackexchange.com/questions/15423/how-to-compute-prediction-bands-for-non-linear-regression References in https://www.mathworks.com/help/stats/nlpredci.html https://stat.ethz.ch/~stahel/courses/cheming/nlreg10E.pdf