Fitting pareto distribution / plotting observed & fitted dists
Some background: I have some data on structural dependencies in a base
of code artifacts. The dependency structure is reflected in terms of
relative node degrees, with each node representing some code unit (just
as an example).
This gives me real data of the following form (sorry for the longish
I collected absolute node degrees of some network structure and computed
the relative node degrees. This is because I want to compare different
networks at another stage.
Based on relative degrees, I attempted some distribution fitting
procedures (see my previous posting). However, I realized that zero
values (representing isolates in the network) in an otherwise continuous
variable are problematic (well, not much of a surprise).
I am wondering how to best deal with those isolates/zeros? Excluding
them entirely would be an option but isolates are relevant for my
analysis. Are there best practises for dealing with such a hybrid variable?
I am aware of censored methods (for regression analysis etc.), but not
when it comes down to distribution fitting, I fear.