embryogrowth (version 6.1.1)

GRTRN_MHmcmc: Metropolis-Hastings algorithm for Embryo Growth Rate Thermal Reaction Norm

Description

Run the Metropolis-Hastings algorithm for data. The number of iterations is n.iter+n.adapt+1 because the initial likelihood is also displayed. I recommend that thin=1 because the method to estimate SE uses resampling. If initial point is maximum likelihood, n.adapt = 0 is a good solution. To get the SE of the point estimates from result_mcmc <- GRTRN_MHmcmc(result=try), use: result_mcmc$SD coda package is necessary for this function. The parameters intermediate and filename are used to save intermediate results every 'intermediate' iterations (for example 1000). Results are saved in a file named filename. The parameter previous is used to indicate the list that has been save using the parameters intermediate and filename. It permits to continue a mcmc search. These options are used to prevent the consequences of computer crash or if the run is very very long and processes with user limited time.

Usage

GRTRN_MHmcmc(result = NULL, n.iter = 10000, parametersMCMC = NULL,
  n.chains = 1, n.adapt = 0, thin = 1, trace = FALSE,
  batchSize = sqrt(n.iter), parallel = TRUE, intermediate = NULL,
  filename = "intermediate.Rdata", previous = NULL)

Arguments

result

An object obtained after a SearchR fit

n.iter

Number of iterations for each step

parametersMCMC

A set of parameters used as initial point for searching with information on priors

n.chains

Number of replicates

n.adapt

Number of iterations before to store outputs

thin

Number of iterations between each stored output

trace

True or False, shows progress

batchSize

Number of observations to include in each batch fo SE estimation

parallel

If true, try to use several cores using parallel computing

intermediate

Period for saving intermediate result, NULL for no save

filename

If intermediate is not NULL, save intermediate result in this file

previous

Previous result to be continued. Can be the filename in which intermediate results are saved.

Value

• A list with resultMCMC being mcmc.list object, resultLnL being likelihoods and parametersMCMC being the parameters used

Details

GRTRN_MHmcmc runs the Metropolis-Hastings algorithm for data (Bayesian MCMC)

Examples

library(embryogrowth)
data(nest)
formated <- FormatNests(nest)
# The initial parameters value can be:
# "T12H", "DHA",  "DHH", "Rho25"
# Or
# "T12L", "T12H", "DHA",  "DHH", "DHL", "Rho25"
x <- structure(c(118.768297442004, 475.750095909406, 306.243694918151,
116.055824800264), .Names = c("DHA", "DHH", "T12H", "Rho25"))
# pfixed <- c(K=82.33) or rK=82.33/39.33
pfixed <- c(rK=2.093313)
resultNest_4p <- searchR(parameters=x, fixed.parameters=pfixed,
	temperatures=formated, derivate=dydt.Gompertz, M0=1.7,
	test=c(Mean=39.33, SD=1.92))
data(resultNest_4p)
pMCMC <- TRN_MHmcmc_p(resultNest_4p, accept=TRUE)
# Take care, it can be very long; several days
result_mcmc_4p <- GRTRN_MHmcmc(result=resultNest_4p,
		parametersMCMC=pMCMC, n.iter=10000, n.chains = 1,
		n.adapt = 0, thin=1, trace=TRUE)
data(result_mcmc_4p)
out <- as.mcmc(result_mcmc_4p)
# This out can be used with coda package
# Test for stationarity and length of chain
require(coda)
heidel.diag(out)
raftery.diag(out)
# plot() can use the direct output of GRTRN_MHmcmc() function.
plot(result_mcmc_4p, parameters=1, xlim=c(0,550))
plot(result_mcmc_4p, parameters=3, xlim=c(290,320))
# summary() permits to get rapidly the standard errors for parameters
# They are store in the result also.
se <- result_mcmc_4p$SD
# the confidence interval is better estimated by:
apply(out[[1]], 2, quantile, probs=c(0.025, 0.975))
# The use of the intermediate method is as followed;
# Here the total mcmc iteration is 10000, but every 1000, intermediate
# results are saved in file intermediate1000.Rdata:
result_mcmc_4p <- GRTRN_MHmcmc(result=resultNest_4p,
parametersMCMC=pMCMC, n.iter=10000, n.chains = 1,
n.adapt = 0, thin=1, trace=TRUE,
intermediate=1000, filename="intermediate1000.Rdata")
# If run has been stopped for any reason, it can be resumed with:
result_mcmc_4p <- GRTRN_MHmcmc(previous="intermediate1000.Rdata")