More Like this

1. Methods for Combining Observational and Experimental Causal Estimates: A Review

   https://doi.org/10.1002/wics.70027  

   Rosenman, Evan (June 2025, WIREs Computational Statistics)

   Gentle, James; Scott, David (Ed.)

   Recent years have seen an explosion in methodological work on combining causal effects estimated from observational and experimental datasets. Observational data have the advantage of being inexpensive and increasingly available from sources such as electronic health records, insurance claims databases, and online learning platforms. These data are representative of target populations, but because treatment assignments are not randomized, they suffer from unmeasured confounding bias. By contrast, as a consequence of randomization, experimental data yield unbiased causal effects. Yet experiments are costly, often involve relatively few units, and may incorporate stringent inclusion criteria that make the studied populations somewhat artificial. A challenge for researchers is how to integrate these two types of data to leverage their respective virtues. Over roughly the past 5 years, many novel approaches have been proposed. As in this review, we restrict our focus to techniques for integrating individual‐level experimental and observational data, without assuming all confounding variables are studied in the observational data. We first “locate” the problem by detailing important considerations from the causal inference and transportability literature. We next discuss three important research traditions that predate modern methodological work: meta‐analysis, Empirical Bayes shrinkage, and historical borrowing. In organizing the growing literature on data‐combination methods, we use a categorization involving five distinct approaches: auxiliary methods, control‐arm augmentation, debiasing, test‐then‐merge, and weighting. Within each category, we summarize recently proposed methodologies, highlighting the strengths and weaknesses of each. We conclude with a discussion of how practitioners might choose between competing approaches when conducting applied work. 

   more » « less
2. Designing experiments informed by observational studies

   https://doi.org/10.1515/jci-2021-0010  

   Rosenman, Evan T.; Owen, Art B. (January 2021, Journal of Causal Inference)

   null (Ed.)

   Abstract The increasing availability of passively observed data has yielded a growing interest in “data fusion” methods, which involve merging data from observational and experimental sources to draw causal conclusions. Such methods often require a precarious tradeoff between the unknown bias in the observational dataset and the often-large variance in the experimental dataset. We propose an alternative approach, which avoids this tradeoff: rather than using observational data for inference, we use it to design a more efficient experiment. We consider the case of a stratified experiment with a binary outcome and suppose pilot estimates for the stratum potential outcome variances can be obtained from the observational study. We extend existing results to generate confidence sets for these variances, while accounting for the possibility of unmeasured confounding. Then, we pose the experimental design problem as a regret minimization problem subject to the constraints imposed by our confidence sets. We show that this problem can be converted into a concave maximization and solved using conventional methods. Finally, we demonstrate the practical utility of our methods using data from the Women’s Health Initiative. 

   more » « less
3. Flexible and Accurate Evaluation of Gravitational-wave Malmquist Bias with Machine Learning

   https://doi.org/10.3847/1538-4357/ac4bc0  

   Talbot, Colm; Thrane, Eric (March 2022, The Astrophysical Journal)

   Abstract Many astronomical surveys are limited by the brightness of the sources, and gravitational-wave searches are no exception. The detectability of gravitational waves from merging binaries is affected by the mass and spin of the constituent compact objects. To perform unbiased inference on the distribution of compact binaries, it is necessary to account for this selection effect, which is known as Malmquist bias. Since systematic error from selection effects grows with the number of events, it will be increasingly important over the coming years to accurately estimate the observational selection function for gravitational-wave astronomy. We employ density estimation methods to accurately and efficiently compute the compact binary coalescence selection function. We introduce a simple pre-processing method, which significantly reduces the complexity of the required machine-learning models. We demonstrate that our method has smaller statistical errors at comparable computational cost than the method currently most widely used allowing us to probe narrower distributions of spin magnitudes. The currently used method leaves 10%–50% of the interesting black hole spin models inaccessible; our new method can probe >99% of the models and has a lower uncertainty for >80% of the models. 

   more » « less
4. Causal Inference With Observational Data and Unobserved Confounding Variables

   https://doi.org/10.1111/ele.70023  

   Byrnes, Jarrett_E K; Dee, Laura E (January 2025, Ecology Letters)

   ABSTRACT Experiments have long been the gold standard for causal inference in Ecology. As Ecology tackles progressively larger problems, however, we are moving beyond the scales at which randomised controlled experiments are feasible. To answer causal questions at scale, we need to also use observational data —something Ecologists tend to view with great scepticism. The major challenge using observational data for causal inference is confounding variables: variables affecting both a causal variable and response of interest. Unmeasured confounders—known or unknown—lead to statistical bias, creating spurious correlations and masking true causal relationships. To combat this omitted variable bias, other disciplines have developed rigorous approaches for causal inference from observational data that flexibly control for broad suites of confounding variables. We show how ecologists can harness some of these methods—causal diagrams to identify confounders coupled with nested sampling and statistical designs—to reduce risks of omitted variable bias. Using an example of estimating warming effects on snails, we show how current methods in Ecology (e.g., mixed models) produce incorrect inferences due to omitted variable bias and how alternative methods can eliminate it, improving causal inferences with weaker assumptions. Our goal is to expand tools for causal inference using observational and imperfect experimental data in Ecology. 

   more » « less
5. A Multiplatform Parallel Approach for Lattice Sieving Algorithms

   https://doi.org/10.1007/978-3-030-60245-1_45  

   Andrzejczak, Michal; Gaj, Kris (October 2020, International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2020)

   Qiu, Meikang (Ed.)

   Lattice sieving is currently the leading class of algorithms for solving the shortest vector problem over lattices. The computational difficulty of this problem is the basis for constructing secure post-quantum public-key cryptosystems based on lattices. In this paper, we present a novel massively parallel approach for solving the shortest vector problem using lattice sieving and hardware acceleration. We combine previously reported algorithms with a proper caching strategy and develop hardware architecture. The main advantage of the proposed approach is eliminating the overhead of the data transfer between a CPU and a hardware accelerator. The authors believe that this is the first such architecture reported in the literature to date and predict to achieve up to 8 times higher throughput when compared to a multi-core high-performance CPU. Presented methods can be adapted for other sieving algorithms hard to implement in FPGAs due to the communication and memory bottleneck. 

   more » « less