Real-time Regression Analysis of Streaming Clustered Data with Possible Abnormal Data Batches

Peter Song (University of Michigan)

In this talk I will introduce an incremental learning algorithm to analyze streaming datasets with correlated outcomes such as longitudinal data and clustered data. We develop a renewable QIF (RenewQIF) method within a paradigm of renewable estimation and incremental inference, in which statistical results are recursively renewed with a current data batch and summary statistics of historical data batches, but with no use of any historical subject-level raw data. We compare our renewable estimation method with both offline QIF method and offline generalized estimating equations (GEE) approach that process the entire cumulative subject-level data all together. We show theoretically and numerically that the RenewQIF enjoys statistical and computational efficiency. In addition, we propose an approach to diagnose the homogeneity assumption of regression coefficients via a sequential goodness-of-fit test as a screening procedure on occurrences of potential abnormal data batches. We implement the proposed methodology by expanding the existing Spark's Lambda architecture for the operation of statistical inference and data quality monitoring. We illustrate the proposed methodology by extensive simulation studies and an analysis of streaming car crash datasets from the National Automotive Sampling System-Crashworthiness Data System (NASS CDS).


Back to Day 1