GraphPad Curve Fitting Guide

How to fit one line to two data sets

How to fit one line to two data sets

Previous topic Next topic No expanding text in this topic  

How to fit one line to two data sets

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

If you entered data for more than one data set on one table and then choose linear or nonlinear regression, Prism usually fits one line or curve through each data set. Follow these steps to fit one line or curve through all the data.

1.Use the nonlinear regression analysis to fit the data, even if you are fitting a straight line. Don't use linear regression.

2.In the first tab of nonlinear regression dialog, pick an equation. If you want to use linear regression, open the "lines" folder of equations and choose "straight line". Fitting this model with nonlinear regression is the same as linear regression, except that Prism offers many more choices in its nonlinear regression dialog.

3.Go to the Constrain tab (of the nonlinear regression dialog), and share all the parameters. If you are fitting a line, share both the slope and also the intercept. "Sharing" means that Prism fits one best-fit value for all the data sets, rather than one for each data set. If you share all the parameters, then Prism fits and draws only one curve or line to all the data.

Caution: One of the assumptions of linear and nonlinear regression is that each data point contributes independent information. This means that it is totally random whether each point is above or below the curve, and by how far. When you fit two data sets to one line or curve, it is likely that you will be violating this assumption. Be careful when interpreting the results.