simu_nbio: Simulation functions

In McCannLab/spatial_fingerprints: Research compendium

Description

Simulation functions

Usage

simu_nbio(
  method = c("lda", "nb", "ml"),
  nrep = 100,
  mxcb = 10,
  nsample = 10,
  ndistr = 20,
  noise = 0,
  pca = FALSE,
  mx_bio = 17,
  ...
)

simu_nbio_order(
  method = c("lda", "nb", "ml"),
  nrep = 100,
  nsample = 10,
  ndistr = 20,
  noise = 0
)

simu_ndistr(
  method = c("lda", "nb", "ml"),
  nrep = 20,
  mxcb = 20,
  nbio = 5,
  nsample = 1,
  noise = 0,
  pca = FALSE
)

simu_nsample(
  method = c("lda", "nb", "ml"),
  nrep = 20,
  mxcb = 20,
  nbio = 5,
  ndistr = 20,
  noise = 0,
  pca = FALSE
)

simu_noise(
  method = c("lda", "nb", "ml"),
  nrep = 20,
  mxcb = 20,
  nbio = 5,
  ndistr = 20,
  nsample = 5,
  pca = FALSE
)

simu_all(
  method = c("lda", "nb", "ml"),
  nrep = 10000,
  mxcb = 700,
  nsample = 1,
  ndistr = 20,
  noise = 0,
  pca = FALSE,
  mx_bio = 17,
  ...
)

Arguments

method	statistical approach, either "lda", "nb" or "ml".
nrep	number of repetition for every iterations.
mxcb	maximum number of bio-tracers combinations (see details).
nsample	number of samples (individuals) to be tested (per region).
ndistr	number of individuals used to generate distributions (per region).
noise	noise to be added.
pca	a logical. See get_data_ready().
mx_bio	maximum bio-tracer number. '
...	further arguments for "ml".
nbio	number of bio-tracers to be used.

Details

If the total number of combinations is smaller than mxcb then the total number of combinations is used.

Functions

• simu_nbio_order: same as simu_nbio() but uses pca and keep the axes ordered.

• simu_ndistr: effect of the number of samples used to build the distribution.

• simu_nsample: effect of the number of samples used for inference.

• simu_noise: effect of the number of samples used for inference.

• simu_all: all combinaison for 1:3

Examples

## Not run: 
simu_nbio("lda")
simu_ndistr("lda")
simu_nsample()

## End(Not run)

McCannLab/spatial_fingerprints documentation built on March 13, 2021, 12:02 a.m.