Calculus: The Language of Change - Epidemic at the Track

The laboratory in this conversation, Boston Mystery, asks you to find the SIR model which best matches the data from the horse epidemic. This is a little difficult with Slinky or any other integrator, so we have a special Java applet to use. It assigns a 'score' to each model (each choice of coefficients) based on how close the curves come to the data points given in the text. The score uses the 'least squares method' described more fully in the second volume of Conversational Calculus, Space, Time, and Infinity. You don't need to know how this is done, but we'll tell you: the vertical distance between the graph and each data point is squared, the squares are added together, then the square-root of the sum is taken. But all you really need to know is that the smaller the score, the better the model fits the data.