Mark van der Laan: Higher order Targeted Maximum Likelihood Estimation
"Higher order Targeted Maximum Likelihood Estimation"
Mark van der Laan, UC Berkeley
Discussant: Alex Luedtke, University of Washington
Abstract: Asymptotic linearity and efficiency of targeted maximum likelihood estimators (TMLE) of target features of the data distribution relies on a a second order remainder being asymptotically negligible. However, in finite samples, the second order remainder can dominate the sampling distribution so that inference based on asymptotic normality would be anti-conservative. We propose a new higher order (say k-th order) TMLE, generalizing the regular (first order) TMLE. We prove that it satisfies an exact linear expansion, in terms of efficient influence functions of sequentially defined higher order fluctuations of the target parameter, with a remainder that is a k+1-th order remainder. As a consequence, this k-th order TMLE allows statistical inference only relying on the k+1-th order remainder being negligible. We present the theoretical result as well as simulations for the second order TMLE for nonparametric estimation of the ATE, and of the integrated squared density.
January 19, 2020
Видео Mark van der Laan: Higher order Targeted Maximum Likelihood Estimation канала Online Causal Inference Seminar
Mark van der Laan, UC Berkeley
Discussant: Alex Luedtke, University of Washington
Abstract: Asymptotic linearity and efficiency of targeted maximum likelihood estimators (TMLE) of target features of the data distribution relies on a a second order remainder being asymptotically negligible. However, in finite samples, the second order remainder can dominate the sampling distribution so that inference based on asymptotic normality would be anti-conservative. We propose a new higher order (say k-th order) TMLE, generalizing the regular (first order) TMLE. We prove that it satisfies an exact linear expansion, in terms of efficient influence functions of sequentially defined higher order fluctuations of the target parameter, with a remainder that is a k+1-th order remainder. As a consequence, this k-th order TMLE allows statistical inference only relying on the k+1-th order remainder being negligible. We present the theoretical result as well as simulations for the second order TMLE for nonparametric estimation of the ATE, and of the integrated squared density.
January 19, 2020
Видео Mark van der Laan: Higher order Targeted Maximum Likelihood Estimation канала Online Causal Inference Seminar
Показать
Комментарии отсутствуют
Информация о видео
22 января 2021 г. 2:55:47
01:04:03
Другие видео канала
![Targeted Learning: From Machine Learning to Inference | Mark van der Laan, PhD | Sep 23, 2020](https://i.ytimg.com/vi/CoO0Y54QxdM/default.jpg)
![Maximum Likelihood, clearly explained!!!](https://i.ytimg.com/vi/XepXtl9YKwc/default.jpg)
![AmirEmad Ghassami: Combining Experimental and Observational Data for Long-Term Causal Effects](https://i.ytimg.com/vi/uVfEo9UuC20/default.jpg)
![Maximum Likelihood Estimation and Bayesian Estimation](https://i.ytimg.com/vi/fjqd0gG-5OE/default.jpg)
![Maximum Likelihood Estimation of the AR(1) Model](https://i.ytimg.com/vi/fxEjTvG24XM/default.jpg)
![Maximum Likelihood - Cramer Rao Lower Bound Intuition](https://i.ytimg.com/vi/i0JiSddCXMM/default.jpg)
![Least Squares as a Maximum Likelihood estimator](https://i.ytimg.com/vi/_-Gnu498s3o/default.jpg)
![Machine learning - Maximum likelihood and linear regression](https://i.ytimg.com/vi/voN8omBe2r4/default.jpg)
![4. Parametric Inference (cont.) and Maximum Likelihood Estimation](https://i.ytimg.com/vi/rLlZpnT02ZU/default.jpg)
![Maximum likelihood estimation of GARCH parameters (FRM T2-26)](https://i.ytimg.com/vi/Xef4LL2KUy0/default.jpg)
![Mireille Schnitzer: Estimation of COVID-19 vaccine effectiveness under the test-negative design](https://i.ytimg.com/vi/NN3V2ggZXjk/default.jpg)
![Method of Moments and Maximum Likelihood Estimation](https://i.ytimg.com/vi/DpzvGq_Rcdw/default.jpg)
![Simulated Maximum Likelihood](https://i.ytimg.com/vi/YO29q7Z43Rw/default.jpg)
![Asymptotic distribution of the maximum likelihood estimator(mle) - finding Fisher information](https://i.ytimg.com/vi/DyLWP5Zx934/default.jpg)
![1. Maximum Likelihood Estimation Basics](https://i.ytimg.com/vi/93fPFOf547Q/default.jpg)
![Maximum Likelihood Examples](https://i.ytimg.com/vi/BFHGIE-nwME/default.jpg)
![The Cramer-Rao Lower Bound: Inference in Maximum Likelihood Estimation](https://i.ytimg.com/vi/igQIsYAlKlY/default.jpg)
![Maximum Likelihood Estimation (MLE) | Score equation | Information | Invariance](https://i.ytimg.com/vi/7kLHJ-F33GI/default.jpg)
![Maximum Likelihood For the Normal Distribution, step-by-step!!!](https://i.ytimg.com/vi/Dn6b9fCIUpM/default.jpg)
![Bin Yu: Predictability, stability, and causality with case study of genetic drivers of heart disease](https://i.ytimg.com/vi/4e2EFrOUGfE/default.jpg)