plot2compare-Bayes.pred-method: Comparing plot method plot2compare for three Bayesian...

In charlottedion/mixedsde: Estimation Methods for Stochastic Differential Mixed Effects Models

Description

Comparison of the results for up to three S4 class Bayes.pred objects

Usage

## S4 method for signature 'Bayes.pred'
plot2compare(x, y, z, newwindow = FALSE,
  plot.legend = TRUE, names, ylim, xlab = "times", ylab = "X", ...)

Arguments

x	Bayes.pred class
y	Bayes.pred class
z	Bayes.pred class (optional)
newwindow	logical(1), if TRUE, a new window is opened for the plot
plot.legend	logical(1), if TRUE, a legend is added
names	character vector with names for the three objects appearing in the legend
ylim	optional
xlab	optional, default 'times'
ylab	optional, default 'X'
...	optional plot parameters

References

Dion, C., Hermann, S. and Samson, A. (2016). Mixedsde: a R package to fit mixed stochastic differential equations.

Examples

random <- 1; sigma <- 0.1; fixed <- 5; param <- c(3, 0.5)
sim <- mixedsde.sim(M = 20, T = 1, N = 50, model = 'OU', random = random, fixed = fixed,
       density.phi = 'normal',param= param, sigma= sigma, X0 = 0, op.plot = 1)

# here: only 100 iterations for example - should be much more!
estim_Bayes_withoutprior <- mixedsde.fit(times = sim$times, X = sim$X, model = 'OU',
             random, estim.method = 'paramBayes',  nMCMC = 100)
prior <- list( m = c(param[1], fixed), v = c(param[1], fixed), alpha.omega = 11,
            beta.omega = param[2]^2*10, alpha.sigma = 10, beta.sigma = sigma^2*9)
estim_Bayes <- mixedsde.fit(times = sim$times, X = sim$X, model = 'OU', random,
           estim.method = 'paramBayes', prior = prior, nMCMC = 100)
plot2compare(estim_Bayes, estim_Bayes_withoutprior, names = c('with prior', 'without prior'))

charlottedion/mixedsde documentation built on May 13, 2019, 3:35 p.m.