"maximum likelihood estimation of beta-normal" & "simulation the gini index of beta-normal distribution" in R
$10-30 USD
Mbyllur
Postuar over 6 years ago
$10-30 USD
Paguhet në dorëzim
Regarding "beta-normal distribution formula" which has been uploaded, I want to debug my scripts to estimate the parameters of the beta-normal distribution.
I've used the maxLik package in r:
------------------------------------------------------------------------------------------------------------------
library(VGAM)
llibrary(VGAM)
library(maxLik)
alfa=4;beta=3;mu=0;sigma=1
n=100
x=rbetanorm(n,alfa,beta,mu,sigma)
logLikFun=function(w){
alfa=w[1]
beta=w[2]
mu=w[3]
sigma=w[4]
z=(x-mu)/sigma
ll={-1*(n*gamma(alfa+beta)-n*log(gamma(alfa))-n*log(gamma(beta))-(n/2)*log(2*pi)-n*log(sigma)+sum((alfa-1)*(pnorm(z,mean=0,sd=1))+(beta-1)*(log(1-pnorm(z,mean=0,sd=1)))-((z^2)/2)))}
ll
}
mle=maxLik(logLikFun,start=c(alfa=5,beta=2.5,mu=1,sigma=1))
summary(mle)
------------------------------------------------------------------------------------------------------------------
the results are quite wrong as shown below:
------------------------------------------------------------------------------------------------------------------
------------------------------------------—
Maximum Likelihood estimation
Newton-Raphson maximisation, 3 iterations
Return code 3: Last step could not find a value above the current.
Boundary of parameter space?
Consider switching to a more robust optimisation method temporarily.
Log-Likelihood: -131157.7
4 free parameters
Estimates:
Estimate Std. error t value Pr(> t)
alfa 5.1037852 NA NA NA
beta 2.3962128 NA NA NA
mu 1.5954536 0.0099872 159.8 <2e-16 *
sigma 0.0444610 0.0001726 257.6 <2e-16 *
—-
Signif. codes: 0 ‘’ 0.001 ‘’ 0.01 ‘’ 0.05 ‘.’ 0.1 ‘ ’ 1
—------------------------------------------
Warning messages:
1: In sqrt(diag(vc)) : NaNs produced
2: In sqrt(diag(vc)) : NaNs produced
------------------------------------------------------------------------------------------------------------------
and I don't know what's the meaning of the fallowing warnings:
> mle=maxLik(logLikFun,start=c(alfa=5,beta=2.5,mu=1,sigma=1))
There were 50 or more warnings (use warnings() to see the first 50)
> summary(mle)
the problem is the NA value for the errors which is not acceptable.
I would be pleased if someone could solve this problem and send me the current scripts which I can edit or add it to my scripts.
More over, regarding "gini index of beta-normal distribution" formula which has been uploaded, I want to simulate this formula by the above method
please inform me about the duration of the project.
ID e Projektit: 15445473
Rreth projektit
Projekt në distancë
Aktive 6 yrs ago
Po kërkoni të fitoni para?
Përfitimet e ofertës për Freelancer
Vendosni buxhetin dhe afatin tuaj
Paguhuni për punën tuaj
Përshkruani propozimin tuaj
Është falas të regjistrohesh dhe të bësh oferta për punë