Simplify multiple simulations — IAM.format

Simplify multiple simulations and group them after computing quantile and median value (per species, fleet, metier, age, year). A selected simulation is kept to illustrate a possible way.

Usage

IAM.format_quant(var_format, probs = c(0.25, 0.75), select_indiv = 1)

Arguments

var_format: Tibble or data.frame format produced by IAM.format.
probs: Quantile arguments for the values distribution once grouped. Vector of 2 numeric values in [0, 1].
select_indiv: Single numeric value which allow to conserve a single simulation value for example. Default is 1.

Value

Format : long format for ggplot and dplyr analysis to facilitate filter and summary computations. If the variable is not defined for a specific dimension (column), this column is filled with NA.

sim_name: Simulation name. chr.
variable: Single value repeated, but is needed if multiple variables are assembled with bind_row(). chr vector
species: Species names, can contain dynamic and static species depending on the variable selected.. chr vector
fleet: Fleet names. fct vector
metier: Metier names. fct vector
age: Ages for dynamic species. fct vector
year: year step for the simulation. dbl vector
quant1: First quantile from probs argument.
quant2: Second quantile from probs argument.
median: Median computed from the grouped values.
value: Value of the variable for every column place indicated on the same row. dbl vector

Author

Maxime Jaunatre

Examples

library(dplyr)
#> 
#> Attaching package: ‘dplyr’
#> The following objects are masked from ‘package:stats’:
#> 
#>     filter, lag
#> The following objects are masked from ‘package:base’:
#> 
#>     intersect, setdiff, setequal, union
library(magrittr)

data("IAM_input_2009")
data("IAM_argum_2009")
sim_statu_quo <- IAM::IAM.model(objArgs = IAM_argum_2009, objInput = IAM_input_2009)
res1 <- IAM.format(sim_statu_quo, c("SSB"), n = 1) %>%
  dplyr::filter(species == "ARC", year < 2012)
res2 <- mutate(res1, n = 2, value = value + rnorm(1, sd = 100))
res <- rbind(res1, res2)

IAM.format_quant(res, c(0.025, 0.975), 2)
#> # A tibble: 3 × 12
#>   sim_name variable species fleet metier age    year quant1 quant2 median  value
#>   <chr>    <chr>    <chr>   <chr> <chr>  <chr> <dbl>  <dbl>  <dbl>  <dbl>  <dbl>
#> 1 NA       SSB      ARC     NA    NA     NA     2009 18723. 18747. 18735. 18747.
#> 2 NA       SSB      ARC     NA    NA     NA     2010 17991. 18014. 18002. 18015.
#> 3 NA       SSB      ARC     NA    NA     NA     2011 17221. 17244. 17232. 17244.
#> # … with 1 more variable: nsim <int>