GraphPad Curve Fitting Guide

Outlier elimination and robust nonlinear regression

Outlier elimination and robust nonlinear regression

Previous topic Next topic No expanding text in this topic  

Outlier elimination and robust nonlinear regression

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

Nonlinear regression, like linear regression, assumes that the scatter of data around the ideal curve follows a Gaussian or normal distribution. Outliers can violate this assumption and invalidate nonlinear regression results. To deal with outliers, Prism offers robust regression and automatic outlier removal.

 

What are the advantages of removing outliers. But isn't it cheating?

Outlier removal has to be used carefully. It must be avoided in many situations.

Once you've identified an outlier, beware or removing it from graphs and analyses. An outlier is not always a 'bad' point.

Prism uses an outlier removal method created by GraphPad Software --the ROUT method. Learn how it works.

When identifying outliers, the step is to fit a curve using a robust method designed so outliers won't affect the curve much. Learn about robust nonlinear regression.