MLFS (version 0.4.2)

predict_ingrowth: predict_ingrowth

Description

ingrowth model for predicting new trees within the MLFS

Usage

predict_ingrowth(
  df_fit,
  df_predict,
  site_vars = site_vars,
  include_climate = include_climate,
  eval_model_ingrowth = TRUE,
  k = 10,
  blocked_cv = TRUE,
  ingrowth_model = "glm",
  rf_mtry = NULL,
  ingrowth_table = NULL,
  DBH_distribution_parameters = NULL
)

Value

a list with four elements:

1. $predicted_ingrowth - a data frame with newly added trees based on the ingrowth predictions

2. $eval_ingrowth - a data frame with predicted and observed number of new trees, separately for each ingrowth level, or character string indicating that ingrowth model was not evaluated

3. $mod_ing_3 - the output model for predicting the ingrowth of trees with code 3

4. $mod_ing_15 - the output model for predicting the ingrowth of trees with code 15 (the output name depends on the code used for this particular ingrowth level)

Arguments

df_fit

a plot-level data with plotID, stand variables and site descriptors, and the two target variables describing the number of ingrowth trees for inner (ingrowth_3) and outer (ingrowth_15) circles

df_predict

data frame which will be used for ingrowth predictions

site_vars

a character vector of variable names which are used as site descriptors

include_climate

logical, should climate variables be included as predictors

eval_model_ingrowth

logical, should the the ingrowth model be evaluated and returned as the output

k

the number of folds to be used in the k fold cross-validation

blocked_cv

logical, should the blocked cross-validation be used in the evaluation phase?

ingrowth_model

model to be used for ingrowth predictions. 'glm' for generalized linear models (Poisson regression), 'ZIF_poiss' for zero inflated Poisson regression and 'rf' for random forest

rf_mtry

a number of variables randomly sampled as candidates at each split of a random forest model for predicting ingrowth. If NULL, default settings are applied.

ingrowth_table

a data frame with 4 variables: (ingrowth) code, DBH_threshold, DBH_max and weight. Ingrowth table is used within the ingrowth sub model to correctly simulate different ingrowth levels and associated upscale weights

DBH_distribution_parameters

A list with deciles of DBH distributions that are used to simulate DBH for new trees, separately for each ingrowth category

Examples

library(MLFS)

data(data_v6)
data(data_ingrowth)
data(ingrowth_table)
data(ingrowth_parameter_list)

ingrowth_outputs <- predict_ingrowth(
   df_fit = data_ingrowth,
   df_predict = data_v6,
   site_vars = c("slope", "elevation", "northness", "siteIndex"),
   include_climate = TRUE,
   eval_model_ingrowth = FALSE,
   rf_mtry = 3,
   k = 10, blocked_cv = TRUE,
   ingrowth_model = 'rf',
   ingrowth_table = ingrowth_table,
   DBH_distribution_parameters = ingrowth_parameter_list)