--- title: "Simulation workflow in bSims" author: "Peter Solymos" output: rmarkdown::html_vignette vignette: > %\VignetteIndexEntry{Simulation workflow in bSims} %\VignetteEngine{knitr::rmarkdown} %\VignetteEncoding{UTF-8} --- ```{r setup,include=FALSE} knitr::opts_chunk$set( collapse = TRUE, comment = "#>" ) par(mar = c(1, 1, 1, 1)) set.seed(429) suppressPackageStartupMessages(library(bSims)) ``` We recommend exploring the simulation settings interactively in the **shiny** apps using `run_app("bsimsH")` app for the homogeneous habitat case and the `run_app("bsimsHER")` app for the stratified habitat case. The apps represent the simulation layers as tabs, the last tab presenting the settings that can be copied onto the clipboard and pasted into the R session or code. In simple situations, comparing results from a few different settings might be enough. Let us consider the following simple comparison: we want to see how much of an effect does roads have when the only effect is that the road stratum is unsuitable. Otherwise there are no behavioral or detectability effects of the road. ```{r sim1} library(bSims) tint <- c(2, 4, 6, 8, 10) rint <- c(0.5, 1, 1.5, 2, Inf) # unlimited ## no road b1 <- bsims_all( road = 0, density = c(1, 1, 0), tint = tint, rint = rint) ## road b2 <- bsims_all( road = 0.5, density = c(1, 1, 0), tint = tint, rint = rint) b1 b2 ``` The `bsims_all` function accepts all the arguments we discussed before for the simulation layers. Unspecified arguments will be taken to be the default value. However, `bsims_all` does not evaluate these arguments, but it creates a closure with the settings. Realizations can be drawn as: ```{r sim2} b1$new() b2$new() ``` Run multiple realizations is done as: ```{r sim3,eval=FALSE} B <- 25 # number of runs bb1 <- b1$replicate(B) bb2 <- b2$replicate(B) ``` The replicate function takes an argument for the number of replicates (`B`) and returns a list of transcript objects with B elements. The `cl` argument can be used to parallelize the work, it can be a numeric value on Unix/Linux/OSX, or a cluster object on any OS. The `recover = TRUE` argument allows to run simulations with error catching. Simulated objects returned by `bsims_all` will contain different realizations and all the conditionally independent layers. Use a customized layered approach if former layers are meant to be kept identical across runs. In more complex situations the **shiny** apps will help identifying corner cases that are used to define a gradient of settings for single or multiple simulation options. You can copy the `bsims_all` settings from the app to be used in simulations. ## Sensitivity analysis In a sensitivity analysis we evaluate how varying one or more settings affect the estimates. This requires setting up a series of values for a setting (argument) while keeping others constant. Let us consider the following scenario: we would like to evaluate how the estimates are changing with increasing road width. We will use the `expand_list` function which creates a list from all combinations of the supplied inputs. Note that we need to wrap vectors inside `list()` to avoid interpreting those as values to iterate over. ```{r grid1} s <- expand_list( road = c(0, 0.5, 1), density = list(c(1, 1, 0)), tint = list(tint), rint = list(rint)) str(s) ``` We now can use this list of settings to run simulations for each. The following illustrates the use of multiple cores: ```{r grid2,eval=FALSE} b <- lapply(s, bsims_all) nc <- 4 # number of cores to use library(parallel) cl <- makeCluster(nc) bb <- lapply(b, function(z) z$replicate(B, cl=cl)) stopCluster(cl) ``` In some cases, we want to evaluate crossed effects of multiple settings. This will give us information about how these settings interact. For example, road width and spatial pattern (random vs. clustered): ```{r grid3} s <- expand_list( road = c(0, 0.5), xy_fun = list( NULL, function(d) exp(-d^2/1^2) + 0.5*(1-exp(-d^2/4^2))), density = list(c(1, 1, 0)), tint = list(tint), rint = list(rint)) str(s) ``` ## Varying landscapes Studying covariate effects on density, cue rates, and detection distances sometimes require that we simulate a series of landscapes that differ. We exploit the fact that arguments to `bsims_all` can be supplied as a list, which is the same as a single-row data frame. This should work for all arguments that accept atomic vectors as arguments: ```{r equal} bsims_all( road = 0.5, density = 1) bsims_all( list( road = 0.5, density = 1)) bsims_all( data.frame( road = 0.5, density = 1)) ``` ```{r variables} # number of stations to visit n <- 5 # random predictors: continuous and discrete x <- data.frame(x1=runif(n,-1,2), x2=rnorm(n)) # density D <- drop(exp(model.matrix(~x2, x) %*% c(0,-0.5))) summary(D) # cue rate phi <- drop(exp(model.matrix(~x1+I(x1^2), x) %*% c(-1,-0.25,-1))) summary(phi) # this data frame collects the columns to be used as arguments s <- data.frame( D=D, vocal_rate = phi, duration = 10, condition = "det1", tau = 1) # each row from s becomes a simulation settings object bb <- lapply(1:n, function(i) bsims_all(s[i,])) # define how you want the data extracted get_counts <- function(b) { o <- b$new() # simulate get_table(o)[1,1] } x$y <- sapply(bb, get_counts) x ``` Read more in the paper: Solymos, P. 2023. Agent-based simulations improve abundance estimation. _Biologia Futura_ 74, 377--392 [DOI 10.1007/s42977-023-00183-2](https://doi.org/10.1007/s42977-023-00183-2).