If you could find physical meaning for a, b, x0/ or a, b, that'd

来源: 2013-02-18 12:00:44 [旧帖] [给我悄悄话] 本文已被阅读:

If you could find physical meaning for a, b, x0/ or a, b, that'd be the best approach.

If you have a hard time coming up with a chemically/physically valid formula, then you can compare some special values of the functions. We should assume x to be non-negative real numbers given the physical meaning of the project. Let's also assume a, b, x0 to be positive, since f is finite and positive. If not, some of the conclusions will be different, but the methodology is the same.

 

1st, x =0: 

when x=0, f =0 only for Hyperbola.

this seems to be physically valid. So sigmoidal doesn't seem to work when x is

close to 0.

2nd, x = infinity,  f should approach an asympotic value, since the quantity of B is limited. This is achieved for both functions.

 

3rd, min/max values for the x in experimental range: 

sigmoidal has a minimum value a/2 when x=x0, is that expected from the experiment? Is that x0 in the range of x you are interested?

Hyperbola is monotonic, with min at x=0, and max at x=infinity.

This would be how I would start comparing the two formulae.