An official website of the United States government
Abstract Genome-wide association studies (GWAS) have identified thousands of genetic variants associated with complex human traits, but only a fraction of variants identified in discovery studies achieve significance in replication studies. Replication in genome-wide association studies has been well-studied in the context of Winner’s Curse, which is the inflation of effect size estimates for significant variants due to statistical chance. However, Winner’s Curse is often not sufficient to explain lack of replication. Another reason why studies fail to replicate is that there are fundamental differences between the discovery and replication studies. A confounding factor can create the appearance of a significant finding while actually being an artifact that will not replicate in future studies. We propose a statistical framework that utilizes genome-wide association studies and replication studies to jointly model Winner’s Curse and study-specific heterogeneity due to confounding factors. We apply this framework to 100 genome-wide association studies from the Human Genome-Wide Association Studies Catalog and observe that there is a large range in the level of estimated confounding. We demonstrate how this framework can be used to distinguish when studies fail to replicate due to statistical noise and when they fail due to confounding.
more »
« less
Inference on Winners
Abstract Policy makers, firms, and researchers often choose among multiple options based on estimates. Sampling error in the estimates used to guide choice leads to a winner’s curse, since we are more likely to select a given option precisely when we overestimate its effectiveness. This winner’s curse biases our estimates for selected options upward and can invalidate conventional confidence intervals. This article develops estimators and confidence intervals that eliminate this winner’s curse. We illustrate our results by studying selection of job-training programs based on estimated earnings effects and selection of neighborhoods based on estimated economic opportunity. We find that our winner’s curse corrections can make an economically significant difference to conclusions but still allow informative inference.
more »
« less
- Award ID(s):
- 1654234
- PAR ID:
- 10484900
- Publisher / Repository:
- Oxford University Press
- Date Published:
- Journal Name:
- The Quarterly Journal of Economics
- Volume:
- 139
- Issue:
- 1
- ISSN:
- 0033-5533
- Format(s):
- Medium: X Size: p. 305-358
- Size(s):
- p. 305-358
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Matise, T (Ed.)Abstract Combining samples for genetic association is standard practice in human genetic analysis of complex traits, but is rarely undertaken in rodent genetics. Here, using 23 phenotypes and genotypes from two independent laboratories, we obtained a sample size of 3076 commercially available outbred mice and identified 70 loci, more than double the number of loci identified in the component studies. Fine-mapping in the combined sample reduced the number of likely causal variants, with a median reduction in set size of 51%, and indicated novel gene associations, including Pnpo, Ttll6, and GM11545 with bone mineral density, and Psmb9 with weight. However, replication at a nominal threshold of 0.05 between the two component studies was low, with less than one-third of loci identified in one study replicated in the second. In addition to overestimates in the effect size in the discovery sample (Winner’s Curse), we also found that heterogeneity between studies explained the poor replication, but the contribution of these two factors varied among traits. Leveraging these observations, we integrated information about replication rates, study-specific heterogeneity, and Winner’s Curse corrected estimates of power to assign variants to one of four confidence levels. Our approach addresses concerns about reproducibility and demonstrates how to obtain robust results from mapping complex traits in any genome-wide association study.more » « less
-
Abstract Resource selection functions (RSFs) are among the most commonly used statistical tools in both basic and applied animal ecology. They are typically parameterized using animal tracking data, and advances in animal tracking technology have led to increasing levels of autocorrelation between locations in such data sets. Because RSFs assume that data are independent and identically distributed, such autocorrelation can cause misleadingly narrow confidence intervals and biased parameter estimates.Data thinning, generalized estimating equations and step selection functions (SSFs) have been suggested as techniques for mitigating the statistical problems posed by autocorrelation, but these approaches have notable limitations that include statistical inefficiency, unclear or arbitrary targets for adequate levels of statistical independence, constraints in input data and (in the case of SSFs) scale‐dependent inference. To remedy these problems, we introduce a method for likelihood weighting of animal locations to mitigate the negative consequences of autocorrelation on RSFs.In this study, we demonstrate that this method weights each observed location in an animal's movement track according to its level of non‐independence, expanding confidence intervals and reducing bias that can arise when there are missing data in the movement track.Ecologists and conservation biologists can use this method to improve the quality of inferences derived from RSFs. We also provide a complete, annotated analytical workflow to help new users apply our method to their own animal tracking data using thectmm Rpackage.more » « less
-
Abstract Instrumental variables have been widely used to estimate the causal effect of a treatment on an outcome. Existing confidence intervals for causal effects based on instrumental variables assume that all of the putative instrumental variables are valid; a valid instrumental variable is a variable that affects the outcome only by affecting the treatment and is not related to unmeasured confounders. However, in practice, some of the putative instrumental variables are likely to be invalid. This paper presents two tools to conduct valid inference and tests in the presence of invalid instruments. First, we propose a simple and general approach to construct confidence intervals based on taking unions of well‐known confidence intervals. Second, we propose a novel test for the null causal effect based on a collider bias. Our two proposals outperform traditional instrumental variable confidence intervals when invalid instruments are present and can also be used as a sensitivity analysis when there is concern that instrumental variables assumptions are violated. The new approach is applied to a Mendelian randomization study on the causal effect of low‐density lipoprotein on globulin levels.more » « less
-
Abstract When estimating a global average treatment effect (GATE) under network interference, units can have widely different relationships to the treatment depending on a combination of the structure of their network neighborhood, the structure of the interference mechanism, and how the treatment was distributed in their neighborhood. In this work, we introduce a sequential procedure to generate and select graph- and treatment-based covariates for GATE estimation under regression adjustment. We show that it is possible to simultaneously achieve low bias and considerably reduce variance with such a procedure. To tackle inferential complications caused by our feature generation and selection process, we introduce a way to construct confidence intervals based on a block bootstrap. We illustrate that our selection procedure and subsequent estimator can achieve good performance in terms of root-mean-square error in several semi-synthetic experiments with Bernoulli designs, comparing favorably to an oracle estimator that takes advantage of regression adjustments for the known underlying interference structure. We apply our method to a real-world experimental dataset with strong evidence of interference and demonstrate that it can estimate the GATE reasonably well without knowing the interference processa priori.more » « less
