R/loocv_thresh_gam.R
loocv_thresh_gam.Rdloocv_thresh_gam applies a LOOCV on a threshold-GAM and its corresponding GAM and
returns TRUE if the threshold-GAM has a lower estimate, else FALSE (see for more infos on
the LOOCV procedure the details section in test_interaction).
loocv_thresh_gam( model, ind_vec, press_vec, t_var_vec, name_t_var, k, a, b, time )
| model | A single GAM object from the model tibble needed to extract the family and the link function. |
|---|---|
| ind_vec | A vector with the IND training observations (including or excluding defined outliers). |
| press_vec | A vector with the training observations (including or excluding defined outliers) of pressure 1 (i.e. the original significant pressure in the GAM(M)). |
| name_t_var | The name of the threshold variable (pressure 2). t_var will be named after this string in the model formula. |
| k | Choice of knots (for the smoothing function |
| a | The lower quantile value of the selected threshold variable, which the estimated threshold is not allowed to exceed; the default is 0.2. |
| b | The upper quantile value of the selected threshold variable, which the estimated threshold is not allowed to exceed; the default is 0.8. |
| time | A vector containing the actual time series. |
The function returns a list with the following 2 sublists:
resultlogical; if TRUE, at least one thresh_gam performs better than its corresponding gam based on LOOCV value.
errorA string capturing potential error messages that occurred as side effects when fitting the threshold GAM for the LOOCV.
thresh_gam which creates a threshold-GAM object and
test_interaction which applies thresh_gam and loocv_thresh_gam
if (FALSE) { # Using the first model in the Baltic Sea demo data loocv_thresh_gam(model = model_gam_ex$model[[1]], ind_vec = ind_init_ex$ind_train[[1]], press_vec = ind_init_ex$press_train[[1]], t_var_vec = ind_init_ex$press_train[[2]], name_t_var = "Swin", k = 4, a = 0.2, b = 0.8, time = ind_init_ex$time_train[[1]]) }