Abstract:Most statistical modeling applications involve the consideration of a candidate collection of models based on various sets of explanatory variables. The candidate models may also differ in terms of the structural formulations for the systematic component and the posited probability distributions for the random component. A common practice is to use an information criterion to select a model from the collection that provides an optimal balance between fidelity to the data and parsimony. The analyst then typically proceeds as if the chosen model was the only model ever considered. However, such a practice fails to account for the variability inherent in the model selection process, which can lead to inappropriate inferential results and conclusions. In recent years, inferential methods have been proposed for multimodel frameworks that attempt to provide an appropriate accounting of modeling uncertainty. In the frequentist paradigm, such methods should ideally involve model selection probabilities, i.e., the relative frequencies of selection for each candidate model based on repeated sampling. Model selection probabilities can be conveniently approximated through bootstrapping. When the Akaike information criterion is employed, Akaike weights are also commonly used as a surrogate for selection probabilities. In this work, we show that the conventional bootstrap approach for approximating model selection probabilities is impacted by bias. We propose a simple correction to adjust for this bias. We also argue that Akaike weights do not provide adequate approximations for selection probabilities, although they do provide a crude gauge of model plausibility.

Bootstrap Inference on Partially Linear Binary Choice Model

Bootstrap Inference for Panel Data Quantile Regression

Bootstrap method-based estimation on the confidence interval for additive interaction in cohort studies

Wild bootstrap inference for penalized quantile regression for longitudinal data

Bootstrap of a Semiparametric Partially Linear Model with Autoregressive Errors

Bootstrap inference in functional linear regression models with scalar response under heteroscedasticity

Bootstrap Approximation of Model Selection Probabilities for Multimodel Inference Frameworks

Empirical-Likelihood-Based Inference for Partially Linear Models

Bootstrap Inference for Fixed‐Effect Models

Wild bootstrap estimation in partially linear models with heteroscedasticity

Parametric Bootstrap Inferences for Panel Data Models

Bootstrap inference for the finite population total under complex sampling designs

Inference In Partially Identified Panel Data Models With Interactive Fixed Effects

Simultaneous Confidence Band for Partially Linear Panel Data Models with Fixed Effects

Estimation and Simultaneous Confidence Bands for Fixed-Effects Panel Data Partially Linear Models

Consistency of the bootstrap for asymptotically linear estimators based on machine learning

Bootstrap Approximation of Wavelet Estimates in a Semiparametric Regression Model

An adaptively resized parametric bootstrap for inference in high-dimensional generalized linear models

Statistical Inference in the Partial Functional Linear Expectile Regression Model

Partial Identification and Inference in Moment Models with Incomplete Data

Assessing Estimation Uncertainty under Model Misspecification