kcde_crossval_estimate_parameter_loss: Using cross-validation, estimate the loss associated with a...

In reichlab/kcde: Kernel conditional density estimation with flexible kernel specifications

Description

Using cross-validation, estimate the loss associated with a particular set of lags and kernel function parameters. We calculate log-score loss

Usage

kcde_crossval_estimate_parameter_loss(combined_params_vector, phi, theta,
  vars_and_offsets, data, leading_rows_to_drop, trailing_rows_to_drop,
  additional_rows_to_drop, kcde_control)

Arguments

combined_params_vector	vector of parameters for filtering and kernel functions that are being estimated
theta	list of kernel function parameters, both those that are being estimated and those that are out of date. Possibly the values of parameters being estimated are out of date; they will be replaced with the values in theta_est_vector.
vars_and_offsets	list representing combinations of variables and lags included in the model
data	the data frame to use in performing cross validation
kcde_control	a list of parameters specifying how the fitting is done

Value

numeric – cross-validation estimate of loss associated with the specified parameters

reichlab/kcde documentation built on May 27, 2019, 4:53 a.m.