Results extracted from REBMIX for the gamma family are much larger than expected

General
1

I am facing a problem, that the results extracted from REBMIX for the gamma family are much larger than expected

This is the R code:

library(mixR)

set.seed(513)

to_shape_rate_gamma(500,700)

to_shape_rate_gamma(2000,700)

myMix500_2 <- rmixgamma(500, c(0.8, 0.2), c(500,2000), c(700,3500))

gamma_2_em = mixfit(myMix500_2, ncomp = 2, family =  "gamma",pi = c(0.3, 0.7), mu = c(400,1500), sd = c(500,1200), tol = 1e-04,max_iter = 500)

gamma_2_em

library(rebmix)

myMix500_2=as.data.frame(myMix500_2)

gamma500est<- REBMIX(Dataset = list(myMix500_2= myMix500_2),
                    Preprocessing = "histogram",
                    cmax = 2,cmin = 2, Criterion = "BIC",
                    pdf = "gamma")

summary(gamma500est)

print(gamma500est)
2

try a larger number of samples, the smaller the sample the harder it is to decide on what mix generated it

rmixgamma(500,

try

rmixgamma(10000,
3

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.