Advice: When to fit a line with nonlinear regression

Linear regression is a special case of nonlinear regression

Linear regression is just a simpler, special, case of nonlinear regression. The calculations are a bit easier (but that only matters to programmers). You can use Prism's nonlinear regression analysis to fit a straight-line model, and the results will be identical to linear regression.

Nonlinear regression offers more options

Using Prism's nonlinear regression analysis to fit a straight line makes sense when you want to:

•	Fit to both a linear and nonlinear model, and compare the two models.

•	Apply differential weighting.

•	Automatically exclude outliers.

•	Use a robust fitting method.

•	Perform a normality test on the residuals.

•	Inspect the correlation matrix or dependencies.

•	Compare the scatter of points from the line with the scatter among replicates with a replicates test.

Nonlinear regression gives more choices if you enter averaged data

If you have replicates at each Y value, you can enter those directly into subcolumns. With both linear and nonlinear regression, Prism will fit the individual replicates unless you ask it to fit the means only.

If you manipulate your data in another program, you may enter your data as Mean, SD (or SEM) and N. In this case, Prism's linear regression analysis fits the means only, ignoring the scatter and sample size. In contrast, Prism's nonlinear regression gives you a choice (in the Weights tab) of fitting just the mean, or of accounting for scatter and sample size. With the latter choice, the results will be identical to what they would have been had you entered the raw data. If you want to account for the SD among replicates, use nonlinear regression.

Some fits that seem linear are really nonlinear

If your Y axis uses a logarithmic or probability scale, then a straight line on the graph is created by a nonlinear model. In this case, although the line on the graph is straight, the model is not actually linear. You need to fit the 'line' with nonlinear regression.

If you want to fit two lines to different segments of the data, this cannot be done with Prism's linear regression analysis. However, Prism's nonlinear regression can fit segmental linear regression.

Using nonlinear regression is no harder than linear regression

Step-by-step instructions.