TAD (version 1.0.0)

launch_analysis_tad: Launch the analysis

Description

Launch distribution analysis

Usage

launch_analysis_tad(
  weights,
  weights_factor,
  trait_data,
  randomization_number,
  aggregation_factor_name = NULL,
  statistics_factor_name = NULL,
  seed = NULL,
  abundance_file = NULL,
  weighted_moments_file = NULL,
  stat_per_obs_file = NULL,
  stat_per_rand_file = NULL,
  stat_skr_param_file = NULL,
  regenerate_abundance_df = FALSE,
  regenerate_weighted_moments_df = FALSE,
  regenerate_stat_per_obs_df = FALSE,
  regenerate_stat_per_rand_df = FALSE,
  regenerate_stat_skr_df = FALSE,
  significativity_threshold = CONSTANTS$DEFAULT_SIGNIFICATIVITY_THRESHOLD,
  lin_mod = CONSTANTS$DEFAULT_LIN_MOD,
  slope_distance = CONSTANTS$DEFAULT_SLOPE_DISTANCE,
  intercept_distance = CONSTANTS$DEFAULT_INTERCEPT_DISTANCE,
  csv_tsv_load_parameters = list()
)

Value

A list of the 9 following named elements:

• raw_abundance_df

• filtered_weights

• filtered_weights_factor

• filtered_trait_data

• abundance_df

• weighted_moments

• statistics_per_observation

• stat_per_rand

• ses_skr

Arguments

weights

the dataframe of weights, one row correspond to a series of observation

weights_factor

the dataframe which contains the different factor linked to the weights

trait_data

a vector of the data linked to the different factor

randomization_number

the number of random abundance matrix to generate

aggregation_factor_name

vector of factor name for the generation of random matrix

statistics_factor_name

vector of factor name for the computation of statistics for each generated matrix

seed

the seed of the pseudo random number generator

abundance_file

the path and name of the RDS file to load/save the dataframe which contains the observed data and the generated matrix

weighted_moments_file

the path and name of the RDS file to load/save the dataframe which contains the calculated moments

stat_per_obs_file

the path and name of the RDS file to load/save the dataframe which contains the statistics for each observed row regarding the random ones

stat_per_rand_file

the path and name of the RDS file to load/save the dataframe which contains the statistics for each random matrix generated

stat_skr_param_file

default=NULL You can provide the output to write the SKR statistics results to.

regenerate_abundance_df

boolean to specify if the abundance dataframe is computed again

regenerate_weighted_moments_df

boolean to specify if the weighted moments dataframe is computed again

regenerate_stat_per_obs_df

boolean to specify if the statistics per observation dataframe is computed again

regenerate_stat_per_rand_df

boolean to specify if the statistics per random matrix dataframe is computed again

regenerate_stat_skr_df

boolean to specify if the stats SKR dataframe is computed again

significativity_threshold

the significance threshold to consider that the observed value is in the randomized value

lin_mod

Indicates the type of linear model to use for (SKR): choose "lm" or "mblm"

slope_distance

slope of the theoretical distribution law (default: slope = 1 intercept = 1.86 skew-uniform distribution family)

intercept_distance

intercept of the theoretical distribution law (default: slope = 1 intercept = 1.86 skew-uniform distribution family)

csv_tsv_load_parameters

a list of parameters for each data structure we want to load. Each element must be named after the data structure we want to load.

Examples

  output_path <- file.path(tempdir(), "outputs")
  dir.create(output_path)
  results <- TAD::launch_analysis_tad(
    weights = TAD::AB[, 5:102],
    weights_factor = TAD::AB[, c("Year", "Plot", "Treatment", "Bloc")],
    trait_data = log(TAD::trait[["SLA"]]),
    aggregation_factor_name = c("Year", "Bloc"),
    statistics_factor_name = (statistics_factor_name <- c("Treatment")),
    regenerate_abundance_df = TRUE,
    regenerate_weighted_moments_df = TRUE,
    regenerate_stat_per_obs_df = TRUE,
    regenerate_stat_per_rand_df = TRUE,
    weighted_moments_file = file.path(output_path, "weighted_moments.csv"),
    stat_per_obs_file = file.path(output_path, "stat_per_obs.csv"),
    stat_per_rand_file = file.path(output_path, "stat_per_rand.csv"),
    stat_skr_param_file = file.path(output_path, "stat_skr_param.csv"),
    randomization_number = 20,
    seed = 1312,
    significativity_threshold = c(0.05, 0.95),
    lin_mod = "lm",
    slope_distance = (
      slope_distance <- TAD::CONSTANTS$SKEW_UNIFORM_SLOPE_DISTANCE
    ),
    intercept_distance = (
      intercept_distance <- TAD::CONSTANTS$SKEW_UNIFORM_INTERCEPT_DISTANCE
    )
  )
  moments_graph <- TAD::moments_graph(
    moments_df = results$weighted_moments,
    statistics_per_observation = results$statistics_per_observation,
    statistics_factor_name = statistics_factor_name,
    statistics_factor_name_breaks = c("Mown_Unfertilized", "Mown_NPK"),
    statistics_factor_name_col = c("#1A85FF", "#D41159"),
    output_path = file.path(output_path, "moments_graph.jpeg"),
    dpi = 100
  )
  skr_graph <- TAD::skr_graph(
    moments_df = results$weighted_moments,
    statistics_factor_name = statistics_factor_name,
    statistics_factor_name_breaks = c("Mown_Unfertilized", "Mown_NPK"),
    statistics_factor_name_col = c("#1A85FF", "#D41159"),
    output_path = file.path(output_path, "skr_graph.jpeg"),
    slope_distance = slope_distance,
    intercept_distance = intercept_distance,
    dpi = 100
  )
  skr_param_graph <- TAD::skr_param_graph(
    skr_param = results$ses_skr,
    statistics_factor_name = statistics_factor_name,
    statistics_factor_name_breaks = c("Mown_Unfertilized", "Mown_NPK"),
    statistics_factor_name_col = c("#1A85FF", "#D41159"),
    slope_distance = slope_distance,
    intercept_distance = intercept_distance,
    save_skr_param_graph = file.path(output_path, "skr_param_graph.jpeg"),
    dpi = 100
  )

  unlink(output_path, recursive = TRUE, force = TRUE)