Skip to contents

Fit a bag of conditional logistic regression/SSF/iSSF models with penalized regression in a train-validate-test setup

Source: R/fit_net_clogit.R, R/fit_net_logit.R
bag_fit_net_functions.Rd

Fits a bag on conditional logistic regressions by calling fit_net_clogit() multiple times to fit different resamples of the input data.

Fits a bag of logistic regressions by calling fit_net_logit() multiple times to fit different resamples of the input data.

Usage

bag_fit_net_clogit(
  f,
  data,
  samples,
  subset_samples = 1:length(samples$train),
  kernel_vars = c("step_length", "ta"),
  metric = c("coxnet.deviance", "Cindex", "conditionalAUC", "conditionalSomersD")[1],
  metrics_evaluate = c("coxnet.deviance", "Cindex", "conditionalAUC"),
  method = c("Lasso", "Ridge", "AdaptiveLasso", "DistanceDecayLasso", "DDLasso",
    "OneZOI-AdaptiveLasso", "OZ-AdaptiveLasso", "Grouped-AdaptiveLasso",
    "G-AdaptiveLasso", "ElasticNet")[1],
  standardize = c("internal", "external", FALSE)[1],
  alpha = NULL,
  penalty.factor = NULL,
  predictor_table = NULL,
  na.action = "na.pass",
  out_dir_file = NULL,
  parallel = c(FALSE, "foreach", "mclapply")[1],
  mc.cores = 2L,
  verbose = FALSE,
  ...
)

bag_fit_net_logit(
  f,
  data,
  samples,
  subset_samples = 1:length(samples$train),
  metric = c(AUC, conditionalBoyce, conditionalSomersD, conditionalAUC)[[1]],
  metrics_evaluate = c("AUC"),
  method = c("Lasso", "Ridge", "AdaptiveLasso", "DistanceDecayLasso", "DDLasso",
    "TruncatedLasso", "OneZOI-AdaptiveLasso", "OZ-AdaptiveLasso",
    "Grouped-AdaptiveLasso", "G-AdaptiveLasso", "ElasticNet")[1],
  standardize = c("internal", "external", FALSE)[1],
  alpha = NULL,
  penalty.factor = NULL,
  predictor_table = NULL,
  na.action = "na.pass",
  out_dir_file = NULL,
  parallel = c(FALSE, "foreach", "mclapply")[1],
  mc.cores = 2L,
  verbose = FALSE,
  ...
)

Arguments

f

[formula]
Formula of the model to be fitted, with all possible candidate terms.

data

[data.frame,tibble]
Complete data set to be analyzed.

samples

[list]
List of samples with at least three elements: train, test, and validate. Typically computed by create_resamples().

subset_samples

[vector]
Vector of sample indices to run (e.g. c(1,2,4,5) or 3:10). By default, all samples in samples are run.

kernel_vars

[character]
Names of movement kernel variables in the model (e.g. "step_length" and "turning_angle").

metric

[function,character]{AUC, conditionalBoyce, conditionalSomersD, conditionalAUC}
Function representing the metric to evaluate goodness-of-fit. One of AUC (Default), conditionalBoyce, conditionalSomersD, and conditionalAUC. A user-defined function might be provided, with a condition that it must be maximized to find the best fit model. It can also be a character, in case it should be one of the following: c("AUC", "conditionalAUC", "conditionalBoyce", "conditionalSomersD").

metrics_evaluate

[character]
Vector of metric names to compute for model evaluation. See description of the argument metric.

method

[character(1)="Lasso"]
Penalized regression method. Default is method = "Lasso", but it could be method = "Ridge", "AdaptiveLasso", "ElasticNet", or different versions of ecology-informed methods. See fit_net_logit() for valid options.

standardize

[character(1)="internal"]{"internal","external",FALSE}
Predictor standardization mode. "internal" uses glmnet's internal standardization; "external" standardizes before calling glmnet; FALSE skips standardization, in case which the standardization should have been done before calling the function to fit the models.

alpha

[numeric(1)=NULL]
Elastic net mixing parameter. If NULL, set automatically by method.

penalty.factor

[numeric]
Penalty factors per coefficient. If NULL, set automatically by method.

predictor_table

[data.frame,NULL]
Predictor table with ZOI metadata, required for distance-decay and ecology-constrained methods. Typically created through [oneimpact::add_zoi_formula()].

na.action

[character(1)="na.pass"]
Action to take for missing values during fitting.

out_dir_file

[character(1)=NULL]
Prefix path for saving individual model results as RDS files.

parallel

[character(1)=FALSE]{"FALSE","foreach","mclapply"}
Parallelization backend. If FALSE (default), no parallelization is made.

mc.cores

[integer(1)=2]
Number of cores for mclapply. If parallel == "foreach", cores must be registered beforehand using parallel::makeCluster() and doParallel::registerDoParallel().

verbose

[logical(1)=FALSE]
Whether to print progress messages during fitting.

...

[any]
Additional options passed to fit_net_logit(), net_logit() and glmnet::glmnet().

Value

A list with:

  • n: total number of models (resamples).

  • formula: model formula.

  • method: fitting method used.

  • metric: metric used for lambda selection.

  • samples: resampling structure.

  • standardize: standardization mode used.

  • models: named list of fitted model objects, one per resample.

  • covariate_mean_sd: data frame of predictor means and SDs used for external standardization (or NULL).

  • numeric_covs: logical vector identifying which covariates are numeric.

A list with:

  • n: total number of models (resamples).

  • formula: model formula.

  • method: fitting method used.

  • metric: metric used for lambda selection.

  • samples: resampling structure.

  • standardize: standardization mode used.

  • models: named list of fitted model objects, one per resample.

  • covariate_mean_sd: data frame of predictor means and SDs used for external standardization (or NULL).

  • numeric_covs: logical vector identifying which covariates are numeric.

See also

fit_net_clogit(), net_clogit(), glmnet::glmnet()

fit_net_logit(), net_logit(), glmnet::glmnet()