Deep neural networks for estimation and inference
From MaRDI portal
(Redirected from Publication:5012758)
Abstract: We study deep neural networks and their use in semiparametric inference. We establish novel rates of convergence for deep feedforward neural nets. Our new rates are sufficiently fast (in some cases minimax optimal) to allow us to establish valid second-step inference after first-step estimation with deep learning, a result also new to the literature. Our estimation rates and semiparametric inference results handle the current standard architecture: fully connected feedforward neural networks (multi-layer perceptrons), with the now-common rectified linear unit activation function and a depth explicitly diverging with the sample size. We discuss other architectures as well, including fixed-width, very deep networks. We establish nonasymptotic bounds for these deep nets for a general class of nonparametric regression-type loss functions, which includes as special cases least squares, logistic regression, and other generalized linear models. We then apply our theory to develop semiparametric inference, focusing on causal parameters for concreteness, such as treatment effects, expected welfare, and decomposition effects. Inference in many other semiparametric contexts can be readily obtained. We demonstrate the effectiveness of deep learning with a Monte Carlo analysis and an empirical application to direct mail marketing.
Recommendations
- Optimal nonparametric inference via deep neural network
- On the rate of convergence of fully connected deep neural network regression estimates
- Nonparametric regression using deep neural networks with ReLU activation function
- Estimation of a regression function on a manifold by fully connected deep neural networks
- On deep learning as a remedy for the curse of dimensionality in nonparametric regression
Cited in
(only showing first 100 items - show all)- Neural nets and implicit inference
- DANTE: deep alternations for training neural networks
- Deep learning for the partially linear Cox model
- First neural conjecturing datasets and experiments
- Optimal nonparametric inference via deep neural network
- Covariate-adjusted Fisher randomization tests for the average treatment effect
- Targeting customers under response-dependent costs
- Estimation of a regression function on a manifold by fully connected deep neural networks
- The costs and benefits of uniformly valid causal inference with high-dimensional nuisance parameters
- scientific article; zbMATH DE number 6500551 (Why is no real title available?)
- scientific article; zbMATH DE number 6928881 (Why is no real title available?)
- Parameter inference with deep jointly informed neural networks
- Adaptive estimation of nonparametric functionals
- Using Machine Learning Methods to Support Causal Inference in Econometrics
- scientific article; zbMATH DE number 7626743 (Why is no real title available?)
- Towards end‐to‐end likelihood‐free inference with convolutional neural networks
- An Algorithmic Interpretation of a Deep Inference System
- Bypassing the Monster: A Faster and Simpler Optimal Algorithm for Contextual Bandits Under Realizability
- Beyond Deep Learning: An Econometric Example
- Deep learning: a statistical viewpoint
- Extremely randomized neural networks for constructing prediction intervals
- DEEP EQUILIBRIUM NETS
- Inference on heterogeneous treatment effects in high‐dimensional dynamic panels under weak dependence
- Change-Plane Analysis for Subgroup Detection with a Continuous Treatment
- Semiparametric estimation of long-term treatment effects
- Asymptotic properties of neural network sieve estimators
- Debiased machine learning of set-identified linear models
- Income risk inequality: Evidence from Spanish administrative records
- Orthogonal statistical learning
- Deep nonparametric regression on approximate manifolds: nonasymptotic error bounds with polynomial prefactors
- Causal inference of general treatment effects using neural networks with a diverging number of confounders
- Asset pricing with neural networks: significance tests
- Assumption-lean falsification tests of rate double-robustness of double-machine-learning estimators
- Incorporating relative error criterion to conformal prediction for positive data
- Calibrating multi-dimensional complex ODE from noisy data via deep neural networks
- Nonlinear Causal Discovery with Confounders
- Deep partial least squares for instrumental variable regression
- Recovering the source term in elliptic equation via deep learning: method and convergence analysis
- An error analysis for deep binary classification with sigmoid loss
- Convergence analysis for over-parameterized deep learning
- Learning optimal biomarker-guided treatment policy for chronic disorders
- How do noise tails impact on deep ReLU networks?
- Double Machine Learning for Sample Selection Models
- Testing Directed Acyclic Graph via Structural, Supervised and Generative Adversarial Learning
- Factor Augmented Sparse Throughput Deep ReLU Neural Networks for High Dimensional Regression
- Doubly Robust Interval Estimation for Optimal Policy Evaluation in Online Learning
- Model-Robust and Efficient Covariate Adjustment for Cluster-Randomized Experiments
- Asymptotic properties of ReLU FFN sieve estimators
- Penalized partially linear regression with deep neural networks for high-dimensional learning
- The FA-SAA algorithm for CVaR optimization
- Deep Regression for Repeated Measurements
- DRM revisited: a complete error analysis
- Learning Green's function efficiently using low-rank approximations
- A nonparametric framework for treatment effect modifier discovery in high dimensions
- Data-rich economic forecasting for actuarial applications
- Predictive density combination using Bayesian machine learning
- Neural networks meet random forests
- Deep learning based residuals in non-linear factor models: precision matrix estimation of returns with low signal-to-noise ratio
- Neural conformal inference for jump diffusion processes
- A Neural Phillips Curve and a Deep Output Gap
- Design-Based Theory for Lasso Adjustment in Randomized Block Experiments and Rerandomized Experiments
- Multiply robust estimation of causal effects using linked data
- Fitting deep neural networks into the statistical regression modelling setting
- Consistency and rate of convergence for deep ReLU neural networks
- A unified framework for covariate adjustment under stratified randomisation
- Convolutional Neural Networks for Valid and Efficient Causal Inference
- The value added of machine learning to causal inference: evidence from revisited studies
- Bayesian neural networks for macroeconomic analysis
- Deep Regression Learning with Optimal Loss Function
- Semi-Supervised Triply Robust Inductive Transfer Learning
- Deep approximate policy iteration
- Extended fiducial inference for individual treatment effects via deep neural networks
- Double robust Bayesian inference on average treatment effects
- De-biased two-sample U-statistics with application to conditional distribution testing
- Semiparametric efficient estimation of genetic relatedness with machine learning methods
- Efficient estimation of single-index models with deep ReQU neural networks
- Should Humans Lie to Machines? The Incentive Compatibility of Lasso and GLM Structured Sparsity Estimators
- Conformal prediction for robust deep nonparametric regression
- Distributionally robust learning for multisource unsupervised domain adaptation
- Handling incomplete outcomes and covariates in cluster-randomized trials: doubly robust estimation, efficiency considerations, and sensitivity analysis
- Causal Machine Learning for Moderation Effects
- Double Debiased Machine Learning Nonparametric Inference with Continuous Treatments
- Selection Bias Adjustment by Functional Transfer Learning via Reproducing Kernel Hilbert Space
- Generative semi-supervised classification
- Change-point detection for multivariate nonparametric regression with deep neural networks
- Nonparametric estimation of conditional average treatment effects under high-dimensional confounding
- Targeted maximum likelihood estimation for integral projection models in population ecology
- Optimal subsampling for high-dimensional partially linear models via machine learning methods
- Integrating random forests into deep neural networks for simultaneous estimation of non-crossing multiple quantiles
- Optimal Neyman-orthogonal subsampling for partially linear mediation models
- Inference on strongly identified functionals of weakly identified functions
- Kolmogorov–Smirnov learning by neural networks with a nonconvex surrogate loss
- Distributional Off-Policy Evaluation with Deep Quantile Process Regression
- Adversarial Estimation of Riesz Representers
- Deep neural network estimation of average treatment effects based on quantile regression
- Causal inference targeting a concentration index for studies of health inequalities
- Binary choice under asymmetric loss in a data-rich environment: theory and an application to algorithmic fairness
- Gradient-based reinforcement learning for dynamic quantile models
- Efficient inference on high-dimensional linear models with missing outcomes
- Supervised Predictive Modeling of High-Dimensional Data with Group ℓ 0 -Norm Constrained Neural Networks
This page was built for publication: Deep neural networks for estimation and inference
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5012758)
