# Models as Approximations, Part I: A Conspiracy of Nonlinearity and Random Regressors in Linear Regression

@inproceedings{Buja2014ModelsAA, title={Models as Approximations, Part I: A Conspiracy of Nonlinearity and Random Regressors in Linear Regression}, author={Andreas Buja and Richard A. Berk and Lawrence D. Brown and Edward Heddy Iii George and Emil Pitkin and Mikhail Traskin and Kaiyun Zhan and Linda H. Zhao}, year={2014} }

More than thirty years ago Halbert White inaugurated a "model-robust" form of statistical inference based on the "sandwich estimator" of standard error. This estimator is known to be "heteroskedasticity-consistent", but it is less well-known to be "nonlinearity-consistent" as well. Nonlinearity, however, raises fundamental issues because regressors are no longer ancillary, hence can't be treated as fixed. The consequences are severe: (1)~the regressor distribution affects the slope parameters… CONTINUE READING

Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

#### Citations

##### Publications citing this paper.

SHOWING 1-10 OF 17 CITATIONS

## Semi-Supervised linear regression

VIEW 11 EXCERPTS

CITES BACKGROUND, RESULTS & METHODS

HIGHLY INFLUENCED

## Graphical Criteria for Efficient Total Effect Estimation via Adjustment in Causal Linear Models

VIEW 2 EXCERPTS

CITES BACKGROUND

HIGHLY INFLUENCED

## Setting the Target: Precise Estimands and the Gap Between Theory and Empirics

VIEW 2 EXCERPTS

CITES BACKGROUND

## Semiparametric Expectile Regression for High-dimensional Heavy-tailed and Heterogeneous Data

VIEW 1 EXCERPT

CITES BACKGROUND

## A Model Free Perspective for Linear Regression: Uniform-in-model Bounds for Post Selection Inference

VIEW 1 EXCERPT

CITES BACKGROUND

## Deterministic Inequalities for Smooth M-estimators.

VIEW 2 EXCERPTS

CITES METHODS

## Evaluating and Improving Targeting Policies with Field Experiments Using Counterfactual Policy Logging

VIEW 2 EXCERPTS

CITES BACKGROUND

## Hierarchical inference for genome-wide association studies: a view on methodology with software

VIEW 1 EXCERPT

CITES BACKGROUND

#### References

##### Publications referenced by this paper.

SHOWING 1-10 OF 51 REFERENCES

## Valid post-selection inference

VIEW 12 EXCERPTS

HIGHLY INFLUENTIAL

## Data Analysis and Approximate Models

VIEW 4 EXCERPTS

HIGHLY INFLUENTIAL

## Optimal M-estimation in high-dimensional regression.

VIEW 5 EXCERPTS

HIGHLY INFLUENTIAL

## Counting the Homeless in Los Angeles County

VIEW 5 EXCERPTS

HIGHLY INFLUENTIAL

## A Theorem of Jacobi and its Generalization

VIEW 5 EXCERPTS

HIGHLY INFLUENTIAL

## Robust Statistics—The Approach Based on Influence Functions

VIEW 5 EXCERPTS

HIGHLY INFLUENTIAL

## Robustness in the Strategy of Scientific Model Building.

VIEW 6 EXCERPTS

HIGHLY INFLUENTIAL

## High-dimensional inference in misspecified linear models

VIEW 3 EXCERPTS

HIGHLY INFLUENTIAL

## Splitting a Predictor at the Upper Quarter or Third and the Lower Quarter or Third

VIEW 6 EXCERPTS

HIGHLY INFLUENTIAL

## On The So-Called “Huber Sandwich Estimator” and “Robust Standard Errors”

VIEW 5 EXCERPTS

HIGHLY INFLUENTIAL