You'll find nonlinear regression most useful if you understand the models you have chosen. The best way to do this is to simulate a curve and then see what happens when you alter a parameter. Viewing graphs of simulated curves is a great way to learn about equations.

Prism can add Gaussian random error to each point in the simulated curve. This can help you test analysis methods. Create a curve with random error, and then analyze the simulated data.

To simulate a curve, start from a data table or graph. Click the Analyze button, select built-in analyses, and then select Simulate Theoretical Curve from the list of curve analyses. Select an equation, enter a value for each parameter, and a range of X values. Check the option box to add random error to each simulated point. Prism generates random errors that follow a Gaussian (bell-shaped) distribution with an SD you enter.

How Prism generates random numbers

Prism can add random values to each of the calculated Y values to simulate experimental error. Prism generates random numbers using routines adapted from Numerical Recipes in C, (W. H. Press et al, second edition, Cambridge Press, 1992; available online at www.nr.com). The function RAN3 (defined in Numerical Recipes) generates uniformly distributed random numbers and the function GASDEV transforms them to a Gaussian distribution with a mean of zero and a standard deviation you enter. Prism uses the time of day when calculating the first random number, so you will get a different series of random numbers every time you run the program.

The only way to generate truly random numbers is through a random physical process such as tossing dice or measuring intervals between radioactive decays. Prism, like all computer programs, generates "random" numbers from defined calculations. Since the sequence of numbers is reproducible, mathematicians say that the numbers are "pseudo-random". The difference between truly random and pseudo-random numbers rarely creates a problem. For most purposes, computer generated random numbers are random enough to simulate data and test analytical methods.