Selecting an equation is the most important step in fitting a curve. The choice cannot be automated.
If your goal is to fit a model in order to understand your data and make comparisons, then choosing a model is a scientific decision that you must make with care. If your goal is to interpolate unknowns from a standard curve, then it matters less which equation you pick, so long as it ends up creating a smooth curve through your data.
•Part of choosing a model is choosing constraints. Don't skip that step. For example, if you choose a sigmoidal dose response model, you must decide whether you wish Prism to find the best fit value of the bottom plateau based on the trend of the data. The alternative is to constrain the bottom plateau to equal zero, if you have subtracted off a baseline, or some other value (defined by controls). A computer can't make these decisions for you. Choosing which constraints to apply to your model is a fundamental decision in data analysis that can have a huge impact on the results.
•If you are fitting several data sets at once, part of choosing a model is deciding which parameters you want to share between data sets. When you share a parameter (a choice on the Constrain tab), Prism finds one best-fit value for the parameter that applies to all the data sets. Read more about shared parameters (global fitting).
Check this option in order to interpolate the concentration of unknown samples from the best-fit curve. Learn more.
With this option, Prism will report the Y value for any X values you enter, and the X value for any Y values you enter (including extrapolating in each direction a distance equal to half the length of the X axis).