lpcde: Estimation and Inference for Local Polynomial Conditional Density Estimators

TL;DR

The paper introduces the R package lpcde, which implements advanced local polynomial methods for estimating and inferring conditional densities and their derivatives, with robust bias correction and boundary adaptivity.

Contribution

It provides a comprehensive software implementation of recent theoretical methods for conditional density estimation, including optimal bandwidth selection and regularization, with practical tools for inference.

Findings

Offers mean square error optimal bandwidth selection

Includes robust bias correction for inference

Demonstrates effectiveness on simulated and real data

Abstract

This paper discusses the R package lpcde, which stands for local polynomial conditional density estimation. It implements the kernel-based local polynomial smoothing methods introduced in Cattaneo, Chandak, Jansson, Ma (2024) for statistical estimation and inference of conditional distributions, densities, and derivatives thereof. The package offers mean square error optimal bandwidth selection and associated point estimators, as well as uncertainty quantification based on robust bias correction both pointwise (e.g., confidence intervals) and uniformly (e.g., confidence bands) over evaluation points. The methods implemented are boundary adaptive whenever the data is compactly supported. The package also implements regularized conditional density estimation methods, ensuring the resulting density estimate is non-negative and integrates to one. We contrast the functionalities of lpcde with existing open-source packages for conditional density estimation, and showcase its main features using simulated and real datasets. An abbreviated version of this article is published in Cattaneo, Chandak, Jansson, Ma (2025 JOSS).