The same method used to interpolate can also be used to extrapolate. But note:
•Prism will extrapolate only a limited distance beyond the data. Usually, it will extrapolate in each direction up to half the range of the X axis.
•To extrapolate further, extend the fit curve to a wider range of X values. For nonlinear regression, use the Range tab to extend the fit curve
•Even if a curve fits the data well, it can be a bit of a stretch to believe that the same model will be accurate when extrapolated beyond the range of the data. It all depends, of course, on the context of the experiment.
•When extrapolating, it is a good idea to ask Prism to report the 95% CI of the interpolated values (which might be very wide).
•When extrapolating, it is a good idea to plot the curve with the 95% confidence bands, so you can see how precise (or imprecise) the curve is as you get further from the data. The curve below is drawn over a range of X values way beyond the data. As the curve gets further from the data, the confidence bands (red) get wider and wider.