May I suggest using tail matching, rather than messing around with the g factor. As long as the rotational correlation time is shorter than the fluorescence decay time, than you will not have the nontrivial hassles associated with determining g factors.
The basic idea is that at long times, t, after excitation the curves I(t)-parallel and I(t)-perpendicular will become identical. Using this information you can scale the two curves, and then determine r(t).
Reference (text): "Time Correlated Single Photon Counting" by Desmond O'Connor & David Philips
Please answer some of my questions regarding the data fitting of anisotropy decays..
1) How do we choose the value of g factor while giving it manually not automatically?
2) can rotation time of the molecule be larger than its fluorescence lifetime?? I think rotations should be complete before the molecule comes down to the ground state??
3) is there any significance of the value of initial residual anisotropy for the lifetime values??