The Use Nelder-Mead Algorithm for Tweedie Maximum Likelihood Estimation


#1

Hello,

I am searching for an example of R code to estimate parameter of a distribution using Nelder-Mead Algorithm, especially for one of the exponential distribution families, that is Tweedie Distribution. Do you have any idea for this one?

Thank you, have a nice day!


#2

Would optim work for you?

With this I think you would need to specify an initial parameter guess and a function that returns the negative (log) likelihood. Something like:

# Initial parameter guess
x0 <- 0.0

neg_log_like <- function(x) {
  ...
}

results <- optim(x0, neg_log_like, method = "Nelder-Mead")

# Print parameter estimate
print(results$par)


#3

Sorry for late replying.. Thank you so much for your help :slight_smile: