combdist.mle {distributionsrd} | R Documentation |
Maximum Likelihood estimation for combined ( single, composite and finite mixture) truncated or complete distributions.
combdist.mle( x, dist, start = NULL, lower = NULL, upper = NULL, components = 1, nested = FALSE, steps = 1, lowertrunc = 0, uppertrunc = Inf, ... )
x |
data vector |
dist |
character vector denoting the distribution(s). |
start |
named numeric vector holding the starting values for the coefficients. |
lower, upper |
Lower and upper bounds to the estimated coefficients, defaults to -Inf and Inf respectively. |
components |
number of components for a mixture distribution. |
nested |
logical indicating whether results should be returned in a nested list or a flat list form, defaults to FALSE. |
steps |
number of steps taken in stepflexmix, defaults to 1. |
lowertrunc, uppertrunc |
lowertrunc- and uppertrunc truncation points, defaults to 0 and Inf respectively |
... |
Additional arguments. |
Returns a named list containing a
Character vector denoting the distributions, separated by an underscore
Nr. of combined distributions
Weights assigned to the respective component distributions
Named vector of coefficients
logical indicator of convergence
Length of the fitted data vector
Nr. of coefficients
x <- rdoubleparetolognormal(1e3) combdist.mle(x = x, dist = "doubleparetolognormal") # Double-Pareto Lognormal combdist.mle(x = x, components = 2, dist = "lnorm", steps = 20) # FMM with 2 components combdist.mle( x = x, dist = c("invpareto", "lnorm", "pareto"), start = c(coeff1.k = 1, coeff2.meanlog = mean(log(x)), coeff2.sdlog = sd(log(x)), coeff3.k = 1), lower = c(1e-10, -Inf, 1e-10, 1e-10), upper = c(Inf, Inf, Inf, Inf), nested = TRUE) # composite distribution