This method constructs an S3 object with associated methods for univariate nonparametric estimation of pdfs, cdfs and quantiles.
hermite_estimator_univar(
N = 50,
standardize = TRUE,
exp_weight_lambda = NA,
observations = c()
)An S3 object of class hermite_estimator_univar, with methods for density function, distribution function and quantile function estimation.
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.
Michael Stephanou <michael.stephanou@gmail.com>
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.