Area calculation of a fitted gaussian curve

Hallo everybody, and thanks in advance for the attention.

I have a vactor of numbers that represent a normalized distribution. Is very easy with the fit commad approximate with one or more gaussian curves the distribution.

Es.

 f = fit(x.',y.','gauss2')
 f =  General model Gauss2:
     f(x) =  a1*exp(-((x-b1)/c1)^2) + a2*exp(-((x-b2)/c2)^2)
     Coefficients (with 95% confidence bounds):
       a1 =     0.03666  (0.02839, 0.04493)
       b1 =       8.072  (7.821, 8.324)
       c1 =      0.7271  (0.5048, 0.9495)
       a2 =     0.06823  (0.06531, 0.07116)
       b2 =       13.26  (12.95, 13.57)
       c2 =       7.753  (7.358, 8.148)

a represent the amplitude, b the average and c the variance. In this way the two groups of coefficients represents two gaussian curves. Now I'd like to calculate the integral or an estimation of it, to know the weigth of each curve on the rappresentation.

Thanks a lot! Emanuele

Answer 1

[Torsten]

http://integrals.wolfram.com/index.jsp?expr=a1*exp%28-%28%28x-b1%29%2Fc1%29%5E2%29+%2B+a2*exp%28-%28%28x-b2%29%2Fc2%29%5E2%29&random=false

Best wishes

Torsten.