Global nonlinear regression is useful in (at least) three situations:
•Test whether a parameter value differs significantly between data sets. Compare the sum-of-squares (to assess goodness-of-fit) when the parameter is shared, with the sum of the sum-of-squares when the parameter is fit individually to each dataset. In Prism, set up this kind of comparison in the Compare tab.
•Fit families of data where each dataset is incomplete, but the entire family of datasets defines the parameters. For example, one data set may do a great job of defining the bottom plateau of a dose-response curve, while another data set defines the top. Fit the two data sets separately, and the results may be ambiguous (very wide confidence intervals). Globally fit both curves, and the results might be very tight. See an example.
•Fit models where the parameter(s) you care about cannot be determined from any one dataset, but only from the relationship between several data sets. Another example is fitting enzyme inhibition data.
The first two uses of global fitting do not require writing special models. The third use requires that you write a model for this purpose.