Predictions from rq.bin Objects

Share:

Description

This function computes predictions based on fitted binary quantile models.

Usage

1
2
3
## S3 method for class 'rq.bin'
predict(object, newdata, na.action = na.pass,
	type = "latent", grid = TRUE, ...)

Arguments

object

an rq.counts object.

newdata

an optional data frame in which to look for variables with which to predict. If omitted, the fitted values are used.

na.action

function determining what should be done with missing values in newdata. The default is to predict NA.

type

the type of prediction required. See details.

grid

either a vector of breakpoints or a logical flag. If TRUE, the breakpoints are set to seq(0.05, 0.95, by = 0.05). If FALSE, the breakpoints are extracted from object$tau. At least 2 breakpoints must be provided.

...

not used.

Details

If type = "latent" (default), the predictions are for the conditional quantiles of the latent response variable; if type = "probability", the predictions are for the conditional probabilities of the observable response. Note that type = "probability" is possible only when normalize = "last" in rq.bin call. The probabilities are predicted over a grid of values.

Value

a vector or a matrix of predictions.

Author(s)

Marco Geraci

See Also

rq.bin, coef.rq.bin

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.