estdweibull3: Estimation of parameters

In DiscreteWeibull: Discrete Weibull Distributions (Type 1 and 3)

Description

Estimation of the parameters of the type 3 discrete Weibull distribution

Usage

estdweibull3(x, method = "P", eps = 1e-04)

Arguments

x	the vector of sample values
method	"ML" for the maximum likelihood method; "M" for the method of moments; "P" for the method of proportions
eps	error threshold for the computation of the moments of the distribution

Value

the vector of the estimates of c and β

Author(s)

Alessandro Barbiero

See Also

ddweibull3

Examples

# Ex1
x <- c(0,0,0,0,0,1,1,1,1,1,1,1,2,2,2,2,3,3,4,6)
estdweibull3(x, "P")
estdweibull3(x, "ML")
estdweibull3(x, "M")
# Ex 2
n <- 20
c <- 1/3
beta <- 2/3
x <- rdweibull3(n, c, beta)
estdweibull3(x, "P")
par <- estdweibull3(x, "ML")
par
-loglikedw3(par, x)
par <- estdweibull3(x, "M")
par
lossdw3(par, x)
n <- 50
x <- rdweibull3(n, c, beta)
estdweibull3(x, "P")
estdweibull3(x, "ML")
estdweibull3(x, "M")
n <- 100
x <- rdweibull3(n, c, beta)
estdweibull3(x, "P")
estdweibull3(x, "ML")
estdweibull3(x, "M")
# Ex 3: a piece of simulation study
nSim <- 50
n <- 50
c <- 0.2
beta <- 0.7
par <- matrix(0, nSim, 2)
for(i in 1:nSim)
{
x <- rdweibull3(n, c, beta)
par[i,] <- estdweibull3(x, "ML")
}
op <- par(mfrow = c(1,2))
boxplot(par[,1], xlab=expression(hat(c)[ML]))
abline(h = c)
boxplot(par[,2], xlab=expression(hat(beta)[ML]))
abline(h = beta)
op <- par()

DiscreteWeibull documentation built on May 2, 2019, 8:58 a.m.