I'm trying to reproduce fitting a beta distribution in Go, given an input of floating-point rates:
rates = [0.20, 0.15, 0.002, 0.017, 0.181, 0.004, ...]
You can do this in R quite easily using fitdistrplus, which chooses reasonable starting values automatically, unlike the MASS package which "struggles":
x <- c(0.0955104277250779, 0.0782381918284555, 0.109683584625186,
0.10115721657354, 0.102377369846524, 0.0691699604743083,
0.0940254652301665, 0.078494747777906, 0.0824474231569216,
0.0886513916653852)
fit <- fitdist(x, "beta")
I am trying to figure out how to fit a beta distribution in Go though which is proving difficult, finding the right functions (stats isn't my strong suit).
I have heard I can use "method-of-moments" as starting values, but also having them automatically be decided would be nice if it works like that out of the box.
