Interpreting a "bad initial values" message
Nonlinear regression works iteratively. Prism starts with initial estimated values for each parameter. It then gradually adjusts these until it converges on the best fit.
In rare circumstances, Prism cannot compute a curve based on the initial parameter values you entered. When it tries, it encounters a math problem. In this case, Prism does not report any results and instead reports "bad initial values".
Math errors that can cause this problem:
- Division by zero
- Taking the logarithm of a negative number
- Taking a negative number to a fractional power.
That third situation requires some comment. In fact, it is mathematically possible sometimes to take a negative number to a fractional power. One example:
-27^(1/3) = -3 ( because -3*-3*-3=-27).
Even though that is mathematically possible, Prism is not able to do that calculation. Prism uses the C language math library, which simply cannot handle any situation where a negative number is taken to a fractional power. You'll need to figure out some way to reorganize the equation.
If you encounter this message, first make sure you entered the right equation. Then check the parameter values (initial values for nonlinear regression) and the range of X values (which you can change in the Range tab of nonlinear regression). The calculation depends on the equation, the parameter values, and the X values, so check all three.