Surprise sampling: improving and extending the local case-control sampling

DOI10.1214/21-EJS1844MaRDI QIDQ2044385zbMATH OpenOpenAlexFDO

Authors Xinwei Shen, Kani Chen, Wen Yu

Publication date 9 August 2021

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2007.02633

zbMATH Keywords

generalized linear modelsHorvitz-Thompson estimatorlocal case-control samplingmodel misspecificationsubsampling

Mathematics Subject Classification ID

Sampling theory, sample surveys (62D05) Generalized linear models (logistic models) (62J12)

Abstract: Fithian and Hastie (2014) proposed a new sampling scheme called local case-control (LCC) sampling that achieves stability and efficiency by utilizing a clever adjustment pertained to the logistic model. It is particularly useful for classification with large and imbalanced data. This paper proposes a more general sampling scheme based on a working principle that data points deserve higher sampling probability if they contain more information or appear "surprising" in the sense of, for example, a large error of pilot prediction or a large absolute score. Compared with the relevant existing sampling schemes, as reported in Fithian and Hastie (2014) and Ai, et al. (2018), the proposed one has several advantages. It adaptively gives out the optimal forms to a variety of objectives, including the LCC and Ai et al. (2018)'s sampling as special cases. Under same model specifications, the proposed estimator also performs no worse than those in the literature. The estimation procedure is valid even if the model is misspecified and/or the pilot estimator is inconsistent or dependent on full data. We present theoretical justifications of the claimed advantages and optimality of the estimation and the sampling design. Different from Ai, et al. (2018), our large sample theory are population-wise rather than data-wise. Moreover, the proposed approach can be applied to unsupervised learning studies, since it essentially only requires a specific loss function and no response-covariate structure of data is needed. Numerical studies are carried out and the evidence in support of the theory is shown.

Recommendations

• Local case-control sampling: efficient subsampling in imbalanced data sets
• Local uncertainty sampling for large-scale multiclass logistic regression
• Estimation for monotone sampling
• Sampling and empirical risk minimization
• Design-unbiased statistical learning in survey sampling

Cites work

• A case-cohort design for epidemiologic cohort studies and disease prevention trials
• A Generalization of Sampling Without Replacement From a Finite Universe
• A statistical perspective on algorithmic leveraging
• Case-cohort and case-control analysis with Cox's model
• End-point sampling
• Generalized case-cohort sampling
• Information-Based Optimal Subdata Selection for Big Data Linear Regression
• Local case-control sampling: efficient subsampling in imbalanced data sets
• Logistic disease incidence models and case-control studies
• Optimal subsampling for large sample logistic regression
• Separate sample logistic discrimination

Cited in

(5)

• Local case-control sampling: efficient subsampling in imbalanced data sets
• Eigen-entropy: a metric for multivariate sampling decisions
• An optimal subsampling design for large-scale Cox model with censored data
• Efficient distributed estimation for expectile regression in increasing dimensions
• Nearly optimal two-step Poisson sampling and empirical likelihood weighting estimation for M-estimation with big data

This page was built for publication: Surprise sampling: improving and extending the local case-control sampling

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2044385)