It depends on the software that you use. Anyway, for a good fit do the following:

1. Most likely you will fit only the imaginary part e'' or M''. Calculate the logarithm of this quantity and fit it with log(HN) to ensure that low values away from the peak are fitted equally well.

2. If your software supports complex numbers, things are easy. Otherwise, you must use the exponential form of complex quantities to calculate powers, e.g. 1+(iwt)^a=1+exp(i*a*pi/2)*(wt)^a = 1 +(cos(a*pi/2)+i*sin(a*pi/2))*(wt)^a = sqrt(1+(wt)^2a+2*cos(a*pi/2)*(wt)^a)*exp(i*atan(sin(a*pi/2)*(wt)^a/(1+cos(a*pi/2)*(wt)^a)))). The rest is easier :)

Please, take a look at the paper:

Use of the dielectric analysis to complement previous thermoanalytical

studies on the system diglycidyl ether of bisphenol

A/1,2 diamine cyclohexane

Lisardo Nunez*, S. Gomez-Barreiro, C.A. Gracia-Fernandez, M.R. Nunez

Polymer 45 (2004) 1167–1175

HEre it is explained how to use the real and imaginary part of the complex modulus (or dielectric complex modulus) to fif the H-N equation

Dear Fernandez and Periklis,

Thank you very much for helping me with your in formations.