Distributions.jl
Distributions.jl copied to clipboard
Feature Request: Fit truncated normal
using Distributions
xs = rand(LogNormal(), 100)
fit(Normal, xs)
fit(truncated(Normal, lb, ub), xs) #ERROR. Want: (mu, sigma, lb, ub)
fit(truncated(Normal, 0., ub), xs) #ERROR. Want: (mu, sigma, ub)
fit(truncated(Normal, 0, Inf), xs) #ERROR. Want: (mu, sigma)
I would also enjoy this feature
Same here