Experimental design issues in big data: the question of bias

DOI10.1007/978-3-030-21140-0_20MaRDI QIDQ3296456zbMATH OpenOpenAlexFDO

Authors Eva Riccomagno, Henry P. Wynn

Publication date 7 July 2020

Published in Statistical Learning of Complex Data (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1712.06916

zbMATH Keywords

experimental designbig dataNash equilibriummodel bias

Mathematics Subject Classification ID

Statistical aspects of big data and data science (62R07) Factorial statistical designs (62K15)

Abstract: Data can be collected in scientific studies via a controlled experiment or passive observation. Big data is often collected in a passive way, e.g. from social media. In studies of causation great efforts are made to guard against bias and hidden confounders or feedback which can destroy the identification of causation by corrupting or omitting counterfactuals (controls). Various solutions of these problems are discussed, including randomization.

Recommendations

• For objective causal inference, design trumps analysis
• Principles of experimental design for big data analysis
• scientific article; zbMATH DE number 1209385
• Combining multiple observational data sources to estimate causal effects
• Observation and experiment. An introduction to causal inference

Cites work

• A Basis for the Selection of a Response Surface Design
• A minimax approach to randomization and estimation in survey sampling
• A minimax approach to sample surveys
• Bayesian inference for causal effects: The role of randomization
• Causality. Models, reasoning, and inference
• Generic identifiability of linear structural equation models by ancestor decomposition
• I-robust and D-robust designs on a finite design space
• Information-Based Optimal Subdata Selection for Big Data Linear Regression
• Learning functions and approximate Bayesian computation design: ABCD
• Maximum Entropy Sampling and Optimal Bayesian Experimental Design
• Minimum bias designs with constraints
• Principles of experimental design for big data analysis
• The central role of the propensity score in observational studies for causal effects

Cited in

(2)

• A panel quantile approach to attrition bias in big data: evidence from a randomized experiment
• Generation of all randomizations using circuits

This page was built for publication: Experimental design issues in big data: the question of bias

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3296456)