CODEWITHSUNDEEP
rlogistic-regressiondistributiondata-fittinggamlss
Fr4nck
Fr4nck

Reputation: 11

R GAMLSS : How to fit a log-logistic distribution (truncated)?

I'm trying to fit a log-logistic distribution (truncated) with GAMLSS and can't find how to do it. In the documentation, it says that for Generalized Beta 2 (GB2) : Setting ν = 1 and τ = 1 in (15.20) gives the log logistic distribution.

To fit the truncated distribution of log-normal, I simply do : gamlss(data~1, family=trun(par=c(Lower,Upper),family="LOGNO",type="both")))

  1. I found that the family "logLO" exists and works but there's no documentation on it and I can't understand the parametrization.

  2. Is it possible to define a new family based on GB2 family by fixing the two parameters ν = 1 and τ = 1 ?

  3. Other ideas ?

Thank you for your help

Franck

Upvotes: 1

Views: 59

Answers (0)

Related Questions