Skip to contents

Predict from SPQR / SPQRX Model

Source: R/SPQRX.R
predict_spqrx.Rd

Generates predictions from a fitted SPQR or SPQRX model object. Supports cumulative distribution (CDF), probability density (PDF), and quantile function (QF) evaluation.

Usage

predict_spqrx(
  object,
  x,
  y = NULL,
  type = "QF",
  tau = 0.5,
  normalize_input = FALSE,
  normalize_output = TRUE
)

Arguments

object

A fitted model object returned by fit_spqrx(), fit.spqrx(), or fit.spqr().

x

Matrix or data frame of covariates for prediction.

y

Optional numeric vector of response values. Required for type = "CDF" and type = "PDF".

type

Character string specifying prediction type: "QF" (quantile function), "CDF", or "PDF".

tau

Numeric value or vector of quantile levels in (0,1). Used when type = "QF".

normalize_input

Logical. If FALSE (default), covariates and response values are normalized using stored model parameters.

normalize_output

Logical. If TRUE (default), quantile predictions are transformed back to the original response scale.

Value

  • For type = "QF": Matrix of predicted quantiles.

  • For type = "CDF": Matrix or vector of cumulative probabilities.

  • For type = "PDF": Matrix or vector of density values.