Online updating method to correct for measurement error in big data streams

• Published in: Computational statistics & data analysis 2020-09, Vol.149, p.106976, Article 106976
• Main Authors: Lee, JooChul, Wang, HaiYing, Schifano, Elizabeth D.
• Format: Article
• Language: English
• Subjects: Data compression; Errors-in-variables; Linear regression; Streaming data
• ISSN: 0167-9473; 1872-7352
• DOI: 10.1016/j.csda.2020.106976

When huge amounts of data arrive in streams, online updating is an important method to alleviate both computational and data storage issues. The scope of previous research for online updating is extended in the context of the classical linear measurement error model. In the case where some covariates are unknowingly measured with error at the beginning of the stream, but then are measured without error after a particular point along the data stream, the updated estimators ignoring the measurement error are biased for the true parameters. Once the covariates measured without error are first observed, a method to correct the bias of the estimators, as well as to correct the biases in their variance estimator, is proposed; after correction, the traditional online updating method can then proceed as usual. Further, asymptotic distributions for the corrected and updated estimators are established. Simulation studies and a real data analysis with an airline on-time dataset are provided to illustrate the performance of the proposed method. •Extends scope of online updating methods to linear measurement error models.•Corrects measurement error biases once measurements are observed precisely.•Establishes asymptotic distributions for the corrected and updated estimators.