Bootstrapping Regression Models

By John Fox. 2002

444 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/a7998de6a01900069ec764e62da84588b026b312

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Models as Approximations I: Consequences Illustrated with Linear Regression (2014, background): “(21) The connection between bootstrap and sandwich estimates is as follows: Proposition 8.”
The Impact of Bootstrap Methods on Time Series Analysis (2003, methodology): “The possibility of drawing an arbitrary resample size was noted as early on as in Bickel and Freedman (1981); nevertheless, the research community focused for some time on the case of resample size n since that choice leads to improved performance, i.e., higher-order accuracy, in the sample mean…”
The Wild Bootstrap, Tamed at Last (2008, unknown): “In order to conduct a fair comparison, we use an improved version of the (y, X) bootstrap suggested by Mammen (1993), and subsequently modified by Flachaire (1999), in which we resample, not the (y,X) pairs as such, but rather regressors (X) and the constrained residuals, transformed according to HC”
Resurrecting weighted least squares (2017, unknown): “In our setting, such inference would correspond to using the WLS estimator in combination with either the pairs bootstrap dating back to Freedman (1981) or the wild bootstrap dating back to Mammen (1993), since these two bootstrap methods are appropriate for regression models that allow for…”
High-dimensional simultaneous inference with the bootstrap (2016, methodology): “4.2 The xyz-paired bootstrap Wemodify here the paired bootstrap for regression (Efron 1979; Liu and Singh 1992) to deal with the case of heteroscedastic errors (Freedman 1981).”
Asymptotic properties of the residual bootstrap for Lasso estimators (2010, methodology): “In the set up of (1.1), the standard bootstrap method is the residual bootstrap of Freedman (1981) .”
Bootstrapping realized multivariate volatility measures (2010, background): “The gains associated with the i.i.d. bootstrap can be quite substantial, especially for the smaller sample sizes, when distortions of the BN-S intervals are larger.”
Asymptotic and Bootstrap Inference for AR(∞) Processes with Conditional Heteroskedasticity (2007, unknown): “This pairwise bootstrap was originally suggested by Freedman (1981) in the cross-sectional context.”
Bootstrapping Lasso Estimators (2011, methodology): “In this situation, the standard approach to bootstrapping is the residual bootstrap (cf. Efron 1979 and Freedman 1981), which was considered by Knight and Fu (2000) in the context of the Lasso estimator.”
On regression adjustments to experimental data (2008, background): “In this example, the models seem to lack adequate substantive foundations, and were somewhat post hoc, as noted by Petitti and Freedman [19].”
1 Bootstrap : A Statistical Method (2008, unknown): “The first two articles dealing with the theory of bootstrap – Bickel and Freedman (1981) and Singh (1981) provided large sample answers for most of the commonly used statistics.”
High-dimensional inference in misspecified linear models (2015, background): “corresponding to the “sandwich formula” in the case with p < n [12, 17, 34, 14].”
A Score Based Approach to Wild Bootstrap Inference (2012, unknown): “Originally proposed as an alternative to the residual bootstrap of Freedman (1981), the wild bootstrap has often been interpreted as a procedure that resamples residuals in a manner that captures any heteroscedasticity in the underlying errors.”
Bootstrapping for multivariate linear regression models (2017, methodology): “Properties of the Mallows metric are developed for random variables on separable Banach spaces of finite dimension [Bickel and Freedman, 1981].”
Thirty Years of Heteroskedasticity-Robust Inference (2013, methodology): “The oldest of the two methods is the pairs bootstrap (Freedman, 1981), in which the investigator resamples from the entire data matrix.”
ON THE CHOICE OF m IN THE m OUT OF n BOOTSTRAP AND CONFIDENCE BOUNDS FOR EXTREMA (2008, methodology): “Case 2: An upper bound for µ, when σ is unknown, is ¯ X n − t n−1,α s n / √ n; the n-bootstrap works (Singh (1981), Bickel and Freedman (1981)), and the tquantile is replaced by the quantile of the bootstrap distribution of √ m( ¯ X * m − ¯ X n )/s * m .”
To Log or Not to Log: Bootstrap as an Alternative to the Parametric Estimation of Moderation Effects in the Presence of Skewed Dependent Variables (2000, background): “Bickel and Freedman (1981; Freedman, 1981) demonstrated that the bootstrap was asymptotically valid for many statistics with known population probability distributions (e.g.,t and OLS regression statistics).”