Tail excess probability prediction using an EQRN_seq object

Usage

EQRN_excess_probability_seq(
  val,
  fit_eqrn,
  X,
  Y,
  intermediate_quantiles,
  interm_lvl = fit_eqrn$interm_lvl,
  crop_predictions = FALSE,
  body_proba = "default",
  proba_type = c("excess", "cdf"),
  seq_len = fit_eqrn$seq_len,
  device = default_device()
)

Arguments

val: Quantile value(s) used to estimate the conditional excess probability or cdf.
fit_eqrn: Fitted "EQRN_seq" object.
X: Matrix of covariates to predict the response's conditional excess probabilities.
Y: Response variable vector corresponding to the rows of X.
intermediate_quantiles: Vector of intermediate conditional quantiles at level fit_eqrn$interm_lvl.
interm_lvl: Optional, checks that interm_lvl == fit_eqrn$interm_lvl.
crop_predictions: Whether to crop out the fist seq_len observations (which are NA) from the returned vector
body_proba: Value to use when the predicted conditional probability is below interm_lvl (in which case it cannot be precisely assessed by the model). If "default" is given (the default), paste0(">",1-interm_lvl) is used if proba_type=="excess", and paste0("<",interm_lvl) is used if proba_type=="cdf".
proba_type: Whether to return the "excess" probability over val (default) or the "cdf" at val.
seq_len: Data sequence length (i.e. number of past observations) used to predict each response quantile. By default, the training fit_eqrn$seq_len is used.
device: (optional) A torch::torch_device(). Defaults to default_device().

Value

Vector of probabilities (and possibly a few body_proba values if val is not large enough) of length nrow(X)

(or nrow(X)-seq_len if crop_predictions).