cross_val_check {reslr}

Cross validation check for spline in time, spline in space time and GAM in order to select the most appropriate number of knots when creating basis functions.

Description

Cross validation check for spline in time, spline in space time and GAM in order to select the most appropriate number of knots when creating basis functions.

Usage

cross_val_check(
  data,
  prediction_grid_res = 50,
  spline_nseg = NULL,
  spline_nseg_t = 20,
  spline_nseg_st = 6,
  n_iterations = 1000,
  n_burnin = 100,
  n_thin = 5,
  n_chains = 2,
  model_type,
  n_fold = 5,
  seed = NULL,
  CI = 0.95
)

Arguments

data

Raw input data

prediction_grid_res

Resolution of grid. Predictions over every 50 years(default) can vary based on user preference, as larger values will reduce computational run time.

spline_nseg

This setting is focused on the Noisy Input Spline model. It provides the number of segments used to create basis functions.

spline_nseg_t

This setting is focused on the Noisy Input Generalised Additive Model. It provides the number of segments used to create basis functions.

spline_nseg_st

This setting is focused on the Noisy Input Generalised Additive Model. It provides the number of segments used to create basis functions.

n_iterations

Number of iterations. Increasing this value will increase the computational run time.

n_burnin

Size of burn-in. This number removes a certain number of samples at the beginning.

n_thin

Amount of thinning.

n_chains

Number of MCMC chains. The number of times the model will be run.

model_type

The user selects their statistical model type. The user can select a Noisy Input Spline in Time using "ni_spline_t". The user can select a Noisy Input Spline in Space Time using "ni_spline_st". The user can select a Noisy Input Generalised Additive Model using "ni_gam_decomp".

n_fold

Number of folds required in the cross validation. The default is 5 fold cross validation.

seed

If the user wants reproducible results, seed stores the output when random selection was used in the creation of the cross validation.

CI

Size of the credible interval required by the user. The default is 0.95 corresponding to 95%.

Value

A list containing the model comparison measures, e.g. Root Mean Square Error (RMSE), and plot of true vs predicted values

Examples

data <- NAACproxydata %>% dplyr::filter(Site == "Cedar Island")
cross_val_check(data = data, model_type = "ni_spline_t",n_fold = 2)