A simple recipe for making accurate parametric inference in finite sample

2019/01/01 by Stéphane Guerrier, Guerrier, Stéphane, Mucyo Karemera +5
Computer Science · Mathematics · #Advanced Statistical Methods and Models #Data Analysis with R #FOS: Computer and information sciences #FOS: Mathematics #Markov Chains and Monte Carlo Methods #Methodology (stat.ME) #Statistical Methods and Inference #Statistics Theory (math.ST)

paper · pdf · doi:10.48550/arxiv.1901.06750

openalex publication_date 2019/01/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Constructing tests or confidence regions that control over the error rates in\nthe long-run is probably one of the most important problem in statistics. Yet,\nthe theoretical justification for most methods in statistics is asymptotic. The\nbootstrap for example, despite its simplicity and its widespread usage, is an\nasymptotic method. There are in general no claim about the exactness of\ninferential procedures in finite sample. In this paper, we propose an\nalternative to the parametric bootstrap. We setup general conditions to\ndemonstrate theoretically that accurate inference can be claimed in finite\nsample.\n

Citations

• The Fitting of Power Series, Meaning Polynomials, Illustrated on Band-Spectroscopic Data
• Robust Estimation of a Location Parameter
• Polar Generation of Random Variates with the t-Distribution
• Bootstrap Methods: Another Look at the Jackknife
• Approximate Bayesian computational methods
• Estimating nonlinear time-series models using simulated vector autoregressions

Related

• Finite Sample Valid Inference via Calibrated Bootstrap
• An accurate percentile method for parametric inference based on asymptotically biased estimators
• Finite sample inference for empirical Bayesian methods
• Bootstrap inference in the presence of bias
• Bootstrapping not under the null?