GraphPad Curve Fitting Guide

"Too few points"

"Too few points"

Previous topic Next topic No expanding text in this topic  

"Too few points"

Previous topic Next topic JavaScript is required for expanding text JavaScript is required for the print function Mail us feedback on this topic!  

Regression requires at least as many data points as parameters. So if you fit a dose-response model with four parameters (Bottom, Top, logEC50, and HillSlope), you must have four or more points or Prism will simply report "too few points".

Having as many points as parameters is the bare minimum. Prism will try to fit the model, and will report a perfect fit, and so not be able to report standard errors or confidence intervals. If there are only a few more points that parameters, Prism is likely to end up with an ambiguous or not converged fit.