plot_weighted {poolABC}

Plot the density estimation of a given parameter

Description

Plots a locfit object obtained after parameter estimation with Approximate Bayesian Computation using the multipleABC() function and merging the multiple posteriors with the mergepost() function.

Usage

plot_weighted(
  prior,
  merged_posterior,
  index,
  limits,
  regWeights = TRUE,
  weighted = TRUE
)

Arguments

prior

is a vector or matrix of simulated parameter values i.e. numbers from the simulations. Each row or vector entry should be a different simulation and each column of a matrix should be a different parameter. This corresponds to the prior distribution and it should contain all the simulated parameter values.

merged_posterior

is a list obtained by the mergepost() function. The output of that function produces a list with the locfit of the various parameters. This function plots those locfits.

index

is an non-negative integer indicating which parameter to plot. It corresponds to the desired entry of the merged_posterior list. So, to plot the first parameter, corresponding to the first entry in the merged_posterior input select 1. To plot the second parameter, select 2 and so on.

limits

is a matrix with two columns and as many rows as there are parameters. Each row should contain the minimum value of the prior for a given parameter in the first column and the maximum value in the second column.

regWeights

logical, indicating whether to plot the posterior density obtained from merging the multiple posteriors with or without the weights of the regression step. The default is TRUE.

weighted

logical, indicating whether to plot the posterior density obtained from merging the multiple posteriors with or without weighting by the overall distance to the global mean. The default is TRUE.

Details

The mergepost() function includes different posterior merging methods and produces locfit objects for each parameter and method. It is possible to select which parameter to plot, with the index input, and whether to plot the density estimation after each accepted point was weighted by its regression weight and by distance of its associated target to the overall mean of the data. If regWeights is set to FALSE, the density estimation obtained without considering the regression weights will be plotted. If weighted is set to FALSE, the density estimation obtained without considering the distance between the mean summary statistics of the target and the mean across all loci.

Value

a plot of the density estimation of a given parameter. This plot will include a title with the name of the parameter. It will also include the density of the prior distribution for that parameter. The density estimation shown here is obtained after merging multiple posteriors for that parameter.

Examples

# load the matrix with parameter values
data(params)
# load the matrix with simulated parameter values
data(sumstats)
# load the matrix with the prior limits
data(limits)

# select some random simulations to act as target just to test the function
targets <- sumstats[c(11:20) ,]
# we should remove those random simulation from the sumstats and params matrices
sumstats <- sumstats[-c(11:20), ]; params <- params[-c(11:20), ]

# parameter estimation for multiple targets
myabc <- multipleABC(targets = targets, params = params, sumstats = sumstats, limits = limits,
tol = 0.01, method = "regression")

# select a random simulation to act as the global value of the summary statistics
# ideally this should be computed from the entirety of the observed data
global <- sumstats[50, ]

# merge the posterior distributions obtained in the previous step
mymerge <- mergepost(target = targets, global = global, post = myabc$adjusted,
wtreg = myabc$weights)

# plot the merged posterior distribution
plot_weighted(prior = params, merged_posterior = mymerge, index = 7, limits = limits)

# note that this is just an example!
# we don't have enough simulations to obtain credible results