predict.LinCDE: predict.LinCDE

In ZijunGao/LinCDE: Conditional Density Estimation via Lindsey's Method and Boosting

predict.LinCDE

Description

This function makes predictions from a LinCDE model.

Usage

## S3 method for class 'LinCDE'
predict(
  object,
  ...,
  X = NULL,
  y = NULL,
  splitPointYTest = NULL,
  densityOnly = TRUE
)

Arguments

object	a LinCDE model.
...	other parameters.
X	input matrix for prediction, of dimension nobs x nvars; each row represents an observation vector.
y	response vector for prediction, of length nobs.
splitPointYTest	vector of response break points. Default is NULL.
densityOnly	a logical value. If TRUE, only return predicted conditional densities density.

Value

The function returns a list of values.

• density: conditional density predictions at y, vector of length nobs. If y = NULL, return NA.

• testLogLikelihood: average test log-likelihood at y. If y = NULL, return NA.

• cellProb: cell probability prediction matrix, of dimension nobs x number of discretization bins; each row represents a cell probability vector.

• yDiscretized: vector of discretized responses/response residuals, of length number of discretization bins. If splitPointYTest = NULL, discretized responses/response residuals from the LinCDE model object are used.

ZijunGao/LinCDE documentation built on Jan. 2, 2023, 11:14 p.m.