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     2.912963   -0.4786992 64.038443 14.64030  1.608921 1.813936
#> 2          2     2.940852   -0.5085643 53.287196 15.10246  1.584533 1.740709
#> 3          3     2.368351   -0.6906364 20.392355 14.00913  1.569154 2.769780
#> 4          4     2.784025   -0.5624740  9.914927 15.61474  1.582818 3.090263
#> 5          5     2.863775   -0.5143786 27.041172 15.34518  1.585760 3.093743
#> 6          6     2.950199   -0.5919163 38.234861 15.99667  1.575896 2.002849
head(gather_sims(m, nsim = 10))
#>   .iteration    .variable   .value
#> 1          1 X.Intercept. 3.273034
#> 2          2 X.Intercept. 2.980942
#> 3          3 X.Intercept. 2.570352
#> 4          4 X.Intercept. 2.934931
#> 5          5 X.Intercept. 3.093088
#> 6          6 X.Intercept. 2.769836
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 with `binwidth`.