Skip to contents

Extract parameter simulations from the joint precision matrix

Source: R/gather-spread.R
gather_sims.Rd

spread_sims() returns a wide-format data frame. gather_sims() returns a long-format data frame. The format matches the format in the tidybayes spread_draws() and gather_draws() functions.

Usage

spread_sims(object, nsim = 200, n_sims = deprecated())

gather_sims(object, nsim = 200, n_sims = deprecated())

Arguments

object

Output from sdmTMB().

nsim

The number of simulation draws.

n_sims

Deprecated: please use nsim.

Value

A data frame. gather_sims() returns a long-format data frame:

  • .iteration: the sample ID

  • .variable: the parameter name

  • .value: the parameter sample value

spread_sims() returns a wide-format data frame:

  • .iteration: the sample ID

  • columns for each parameter with a sample per row

Examples

m <- sdmTMB(density ~ depth_scaled,
  data = pcod_2011, mesh = pcod_mesh_2011, family = tweedie())
head(spread_sims(m, nsim = 10))
#>   .iteration X.Intercept. depth_scaled    range      phi tweedie_p  sigma_O
#> 1          1     2.191667   -0.5901819 31.85191 16.15949  1.566081 2.430088
#> 2          2     2.602605   -0.7367376 16.97427 16.47078  1.598116 3.120594
#> 3          3     2.690644   -1.0831837 42.99310 15.04725  1.607441 1.835503
#> 4          4     2.810274   -0.6135098 28.87011 15.27632  1.600637 2.412999
#> 5          5     2.943655   -0.6800500 37.60646 15.80146  1.595988 2.410816
#> 6          6     3.214597   -0.7106337 48.36184 15.35741  1.613159 1.748936
head(gather_sims(m, nsim = 10))
#>   .iteration    .variable   .value
#> 1          1 X.Intercept. 3.428553
#> 2          2 X.Intercept. 2.842100
#> 3          3 X.Intercept. 2.931986
#> 4          4 X.Intercept. 2.816825
#> 5          5 X.Intercept. 2.101858
#> 6          6 X.Intercept. 2.805431
samps <- gather_sims(m, nsim = 1000)

if (require("ggplot2", quietly = TRUE)) {
  ggplot(samps, aes(.value)) + geom_histogram() +
    facet_wrap(~.variable, scales = "free_x")
}
#> `stat_bin()` using `bins = 30`. Pick better value `binwidth`.