hermite_estimator_univar: A class to sequentially estimate univariate pdfs, cdfs and...

View source: R/hermite_estimator_univar.R

hermite_estimator_univarR Documentation

A class to sequentially estimate univariate pdfs, cdfs and quantile functions


This method constructs an S3 object with associated methods for univariate nonparametric estimation of pdfs, cdfs and quantiles.


  N = 50,
  standardize = TRUE,
  exp_weight_lambda = NA,
  observations = c()



An integer between 0 and 75. The upper bound has been chosen as a value that yields an estimator that is reasonably fast and that remains robust to numerical issues. The Hermite series based estimator is truncated at N+1 terms.


A boolean value. Determines whether the observations are standardized, a transformation which often improves performance.


A numerical value between 0 and 1. This parameter controls the exponential weighting of the Hermite series based estimator. If this parameter is NA, no exponential weighting is applied.


A numeric vector. A vector of observations to be incorporated into the estimator.


The hermite_estimator_univar class allows the sequential or one-pass batch estimation of the full probability density function, cumulative distribution function and quantile function. It is well suited to streaming data (both stationary and non-stationary) and to efficient estimation in the context of massive or distributed data sets. Indeed, estimators constructed on different subsets of a distributed data set can be consistently merged.


An S3 object of class hermite_estimator_univar, with methods for density function, distribution function and quantile function estimation.


Michael Stephanou <michael.stephanou@gmail.com>

hermiter documentation built on May 31, 2023, 6:30 p.m.