Year: 2018
Type: preprint
Abstract: Estimators based on influence functions (IFs) have been shown to be effective in many settings, especially when combined with machine learning techniques. By focusing on estimating a specific target of interest (e.g., the average effect of a treatment), rather than on estimating the full underlying data generating distribution, IF-based estimators are often able to achieve asymptotically optimal mean-squared error. Still, many researchers find IF-based estimators to be opaque or overly technical, which makes their use less prevalent and their benefits less available. To help foster understanding and trust in IF-based estimators, we present tangible, visual illustrations of when and how IF-based estimators can outperform standard ``plug-in'' estimators. The figures we show are based on connections between IFs, gradients, linear approximations, and Newton-Raphson.
Source: arXiv (Cornell University)
Authors: Fisher, Aaron, Kennedy, Edward H.
Institution: Carnegie Mellon University
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
Awards: G1435456691, G1485023119, Relationship Between Multiple Environmental Exposures and CVD Incidence and Survival: Vulnerability and Susceptibility, Casual Inference with Interference for Evaluating Air Quality Policies, Optimal Nonparametric Estimation of High-Dimensional Functionals in Causal Inference, Statistical Informatics for Cancer Research, Bayesian Methods for Comparative Effectiveness Research with Observational Data, G8956842828, G9000018821, G9000039249, G9000062543, G9000114572, G9000165177, G921101280
OpenAlex ID: W4289422682