Voilà, trois exemples. J'ai rendu le code beaucoup moins efficace qu'il ne le serait dans une application réelle afin de rendre la logique plus claire (j'espère.)
# We'll assume estimation of a Poisson mean as a function of x
x <- runif(100)
y <- rpois(100,5*x) # beta = 5 where mean(y[i]) = beta*x[i]
# Prior distribution on log(beta): t(5) with mean 2
# (Very spread out on original scale; median = 7.4, roughly)
log_prior <- function(log_beta) dt(log_beta-2, 5, log=TRUE)
# Log likelihood
log_lik <- function(log_beta, y, x) sum(dpois(y, exp(log_beta)*x, log=TRUE))
# Random Walk Metropolis-Hastings
# Proposal is centered at the current value of the parameter
rw_proposal <- function(current) rnorm(1, current, 0.25)
rw_p_proposal_given_current <- function(proposal, current) dnorm(proposal, current, 0.25, log=TRUE)
rw_p_current_given_proposal <- function(current, proposal) dnorm(current, proposal, 0.25, log=TRUE)
rw_alpha <- function(proposal, current) {
# Due to the structure of the rw proposal distribution, the rw_p_proposal_given_current and
# rw_p_current_given_proposal terms cancel out, so we don't need to include them - although
# logically they are still there: p(prop|curr) = p(curr|prop) for all curr, prop
exp(log_lik(proposal, y, x) + log_prior(proposal) - log_lik(current, y, x) - log_prior(current))
}
# Independent Metropolis-Hastings
# Note: the proposal is independent of the current value (hence the name), but I maintain the
# parameterization of the functions anyway. The proposal is not ignorable any more
# when calculation the acceptance probability, as p(curr|prop) != p(prop|curr) in general.
ind_proposal <- function(current) rnorm(1, 2, 1)
ind_p_proposal_given_current <- function(proposal, current) dnorm(proposal, 2, 1, log=TRUE)
ind_p_current_given_proposal <- function(current, proposal) dnorm(current, 2, 1, log=TRUE)
ind_alpha <- function(proposal, current) {
exp(log_lik(proposal, y, x) + log_prior(proposal) + ind_p_current_given_proposal(current, proposal)
- log_lik(current, y, x) - log_prior(current) - ind_p_proposal_given_current(proposal, current))
}
# Vanilla Metropolis-Hastings - the independence sampler would do here, but I'll add something
# else for the proposal distribution; a Normal(current, 0.1+abs(current)/5) - symmetric but with a different
# scale depending upon location, so can't ignore the proposal distribution when calculating alpha as
# p(prop|curr) != p(curr|prop) in general
van_proposal <- function(current) rnorm(1, current, 0.1+abs(current)/5)
van_p_proposal_given_current <- function(proposal, current) dnorm(proposal, current, 0.1+abs(current)/5, log=TRUE)
van_p_current_given_proposal <- function(current, proposal) dnorm(current, proposal, 0.1+abs(proposal)/5, log=TRUE)
van_alpha <- function(proposal, current) {
exp(log_lik(proposal, y, x) + log_prior(proposal) + ind_p_current_given_proposal(current, proposal)
- log_lik(current, y, x) - log_prior(current) - ind_p_proposal_given_current(proposal, current))
}
# Generate the chain
values <- rep(0, 10000)
u <- runif(length(values))
naccept <- 0
current <- 1 # Initial value
propfunc <- van_proposal # Substitute ind_proposal or rw_proposal here
alphafunc <- van_alpha # Substitute ind_alpha or rw_alpha here
for (i in 1:length(values)) {
proposal <- propfunc(current)
alpha <- alphafunc(proposal, current)
if (u[i] < alpha) {
values[i] <- exp(proposal)
current <- proposal
naccept <- naccept + 1
} else {
values[i] <- exp(current)
}
}
naccept / length(values)
summary(values)
Pour l'échantillonneur vanille, on obtient:
> naccept / length(values)
[1] 0.1737
> summary(values)
Min. 1st Qu. Median Mean 3rd Qu. Max.
2.843 5.153 5.388 5.378 5.594 6.628
ce qui est une faible probabilité d'acceptation, mais quand même ... régler la proposition aiderait ici, ou en adopter une autre. Voici les résultats de la proposition de marche aléatoire:
> naccept / length(values)
[1] 0.2902
> summary(values)
Min. 1st Qu. Median Mean 3rd Qu. Max.
2.718 5.147 5.369 5.370 5.584 6.781
Des résultats similaires, comme on pourrait l'espérer, et une meilleure probabilité d'acceptation (visant ~ 50% avec un paramètre.)
Et, pour être complet, l'échantillonneur d'indépendance:
> naccept / length(values)
[1] 0.0684
> summary(values)
Min. 1st Qu. Median Mean 3rd Qu. Max.
3.990 5.162 5.391 5.380 5.577 8.802
Parce qu'il ne "s'adapte" pas à la forme de la partie postérieure, il a tendance à avoir la probabilité d'acceptation la plus faible et est le plus difficile à régler correctement pour ce problème.
Notez que d'une manière générale, nous préférerions des propositions avec des queues plus grosses, mais c'est un tout autre sujet.