arXiv subjects

Andrew Gelman

Publications and source records attributed to Andrew Gelman.

At least 55 records · Page 3Linked to original sources

arXiv2021

To design interfaces for exploratory data analysis, we need theories of graphical inference

Research and development in computer science and statistics have produced increasingly sophisticated software interfaces for interactive and exploratory analysis, optimized for easy pattern finding and data exposure. But design philosophies that emphasize exploration over other phases of analysis risk confusing a need for flexibility with a conclusion that exploratory visual analysis is inherently model-free and cannot be formalized. We describe how without a grounding in theories of human statistical inference, research in exploratory visual analysis can lead to contradictory interface objectives and representations of uncertainty that can discourage users from drawing valid inferences. We discuss how the concept of a model check in a Bayesian statistical framework unites exploratory and confirmatory analysis, and how this understanding relates to other proposed theories of graphical inference. Viewing interactive analysis as driven by model checks suggests new directions for software and empirical research around exploratory and visual analysis. For example, systems should enable specifying and explicitly comparing data to null and other reference distributions and better representations of uncertainty. Implications of Bayesian and other theories of graphical inference should be tested against outcomes of interactive analysis by people to drive theory development.

cs.HC↗

arXiv2021

Rank-normalization, folding, and localization: An improved $\widehat{R}$ for assessing convergence of MCMC

Markov chain Monte Carlo is a key computational tool in Bayesian statistics, but it can be challenging to monitor the convergence of an iterative stochastic algorithm. In this paper we show that the convergence diagnostic $\widehat{R}$ of Gelman and Rubin (1992) has serious flaws. Traditional $\widehat{R}$ will fail to correctly diagnose convergence failures when the chain has a heavy tail or when the variance varies across the chains. In this paper we propose an alternative rank-based diagnostic that fixes these problems. We also introduce a collection of quantile-based local efficiency measures, along with a practical approach for computing Monte Carlo error estimates for quantiles. We suggest that common trace plots should be replaced with rank plots from multiple chains. Finally, we give recommendations for how these methods should be used in practice.

stat.CO↗

arXiv2021

What are the most important statistical ideas of the past 50 years?

We review the most important statistical ideas of the past half century, which we categorize as: counterfactual causal inference, bootstrapping and simulation-based inference, overparameterized models and regularization, Bayesian multilevel models, generic computation algorithms, adaptive decision analysis, robust inference, and exploratory data analysis. We discuss key contributions in these subfields, how they relate to modern computing and big data, and how they might be developed and extended in future decades. The goal of this article is to provoke thought and discussion regarding the larger themes of research in statistics and data science.

stat.ME↗

arXiv2021

Bayesian hierarchical stacking: Some models are (somewhere) useful

Stacking is a widely used model averaging technique that asymptotically yields optimal predictions among linear averages. We show that stacking is most effective when model predictive performance is heterogeneous in inputs, and we can further improve the stacked mixture with a hierarchical model. We generalize stacking to Bayesian hierarchical stacking. The model weights are varying as a function of data, partially-pooled, and inferred using Bayesian inference. We further incorporate discrete and continuous inputs, other structured priors, and time series and longitudinal data. To verify the performance gain of the proposed method, we derive theory bounds, and demonstrate on several applied problems.

stat.ME↗

arXiv2021

Inference from Non-Random Samples Using Bayesian Machine Learning

We consider inference from non-random samples in data-rich settings where high-dimensional auxiliary information is available both in the sample and the target population, with survey inference being a special case. We propose a regularized prediction approach that predicts the outcomes in the population using a large number of auxiliary variables such that the ignorability assumption is reasonable while the Bayesian framework is straightforward for quantification of uncertainty. Besides the auxiliary variables, inspired by Little & An (2004), we also extend the approach by estimating the propensity score for a unit to be included in the sample and also including it as a predictor in the machine learning models. We show through simulation studies that the regularized predictions using soft Bayesian additive regression trees yield valid inference for the population means and coverage rates close to the nominal levels. We demonstrate the application of the proposed methods using two different real data applications, one in a survey and one in an epidemiology study.

stat.ME↗

arXiv2021

Routine Hospital-based SARS-CoV-2 Testing Outperforms State-based Data in Predicting Clinical Burden

Throughout the COVID-19 pandemic, government policy and healthcare implementation responses have been guided by reported positivity rates and counts of positive cases in the community. The selection bias of these data calls into question their validity as measures of the actual viral incidence in the community and as predictors of clinical burden. In the absence of any successful public or academic campaign for comprehensive or random testing, we have developed a proxy method for synthetic random sampling, based on viral RNA testing of patients who present for elective procedures within a hospital system. We present here an approach under multilevel regression and poststratification (MRP) to collecting and analyzing data on viral exposure among patients in a hospital system and performing statistical adjustment that has been made publicly available to estimate true viral incidence and trends in the community. We apply our MRP method to track viral behavior in a mixed urban-suburban-rural setting in Indiana. This method can be easily implemented in a wide variety of hospital settings. Finally, we provide evidence that this model predicts the clinical burden of SARS-CoV-2 earlier and more accurately than currently accepted metrics.

stat.AP↗

arXiv2021

Making the most of imprecise measurements: Changing patterns of arsenic concentrations in shallow wells of Bangladesh from laboratory and field data

Millions of people in Bangladesh drink well water contaminated with arsenic. Despite the severity of this heath crisis, little is known about the extent to which groundwater arsenic concentrations change over time: Are concentrations generally rising, or is arsenic being flushed out of aquifers? Are spatially patterns of high and low concentrations across wells homogenizing over time, or are these spatial gradients becoming more pronounced? To address these questions, we analyze a large set of arsenic concentrations that were sampled within a 25 km$^2$ area of Bangladesh over time. We compare two blanket survey collected in 2000/2001 and 2012/2013 from the same villages but relying on a largely different set of wells. The early set consists of 4574 accurate laboratory measurements, but the later set poses a challenge for analysis because it is composed of 8229 less accurate categorical measurements conducted in the field with a kit. We construct a Bayesian model that jointly calibrates the measurement errors, applies spatial smoothing, and describes the spatiotemporal dynamic with a diffusion-like process model. Our statistical analysis reveals that arsenic concentrations change over time and that their mean dropped from 110 to 96 $μ$g/L over 12 years, although one quarter of individual wells are inferred to see an increase. The largest decreases occurred at the wells with locally high concentrations where the estimated Laplacian indicated that the arsenic surface was strongly concave. However, well with initially low concentrations were unlikely to be contaminated by nearby high concentration wells over a decade. We validate the model using a posterior predictive check on an external subset of laboratory measurements from the same 271 wells in the same study area available for 2000, 2014, and 2015.

stat.AP↗

arXiv2020

A proposal for informative default priors scaled by the standard error of estimates

If we have an unbiased estimate of some parameter of interest, then its absolute value is positively biased for the absolute value of the parameter. This bias is large when the signal-to-noise ratio (SNR) is small, and it becomes even larger when we condition on statistical significance; the winner's curse. This is a frequentist motivation for regularization. To determine a suitable amount of shrinkage, we propose to estimate the distribution of the SNR from a large collection or corpus of similar studies and use this as a prior distribution. The wider the scope of the corpus, the less informative the prior, but a wider scope does not necessarily result in a more diffuse prior. We show that the estimation of the prior simplifies if we require that posterior inference is equivariant under linear transformations of the data. We demonstrate our approach with corpora of 86 replication studies from psychology and 178 phase 3 clinical trials. Our suggestion is not intended to be a replacement for a prior based on full information about a particular problem; rather, it represents a familywise choice that should yield better long-term properties than the current default uniform prior, which has led to systematic overestimates of effect sizes and a replication crisis when these inflated estimates have not shown up in later studies.

stat.ME↗

arXiv2020

Holes in Bayesian Statistics

Every philosophy has holes, and it is the responsibility of proponents of a philosophy to point out these problems. Here are a few holes in Bayesian data analysis: (1) the usual rules of conditional probability fail in the quantum realm, (2) flat or weak priors lead to terrible inferences about things we care about, (3) subjective priors are incoherent, (4) Bayesian decision picks the wrong model, (5) Bayes factors fail in the presence of flat or weak priors, (6) for Cantorian reasons we need to check our models, but this destroys the coherence of Bayesian inference. Some of the problems of Bayesian statistics arise from people trying to do things they shouldn't be trying to do, but other holes are not so easily patched. In particular, it may be a good idea to avoid flat, weak, or conventional priors, but such advice, if followed, would go against the vast majority of Bayesian practice and requires us to confront the fundamental incoherence of Bayesian inference. This does not mean that we think Bayesian inference is a bad idea, but it does mean that there is a tension between Bayesian logic and Bayesian workflow which we believe can only be resolved by considering Bayesian logic as a tool, a way of revealing inevitable misfits and incoherences in our model assumptions, rather than as an end in itself.

math.ST↗

arXiv2020

A Fast Linear Regression via SVD and Marginalization

We describe a numerical scheme for evaluating the posterior moments of Bayesian linear regression models with partial pooling of the coefficients. The principal analytical tool of the evaluation is a change of basis from coefficient space to the space of singular vectors of the matrix of predictors. After this change of basis and an analytical integration, we reduce the problem of finding moments of a density over k + m dimensions, to finding moments of an m-dimensional density, where k is the number of coefficients and k + m is the dimension of the posterior. Moments can then be computed using, for example, MCMC, the trapezoid rule, or adaptive Gaussian quadrature. An evaluation of the SVD of the matrix of predictors is the dominant computational cost and is performed once during the precomputation stage. We demonstrate numerical results of the algorithm. The scheme described in this paper generalizes naturally to multilevel and multi-group hierarchical regression models where normal-normal parameters appear.

stat.CO↗

arXiv2020

Bayesian Workflow

The Bayesian approach to data analysis provides a powerful way to handle uncertainty in all observations, model parameters, and model structure using probability theory. Probabilistic programming languages make it easier to specify and fit Bayesian models, but this still leaves us with many options regarding constructing, evaluating, and using these models, along with many remaining challenges in computation. Using Bayesian inference to solve real-world problems requires not only statistical skills, subject matter knowledge, and programming, but also awareness of the decisions made in the process of data analysis. All of these aspects can be understood as part of a tangled workflow of applied Bayesian statistics. Beyond inference, the workflow also includes iterative model building, model checking, validation and troubleshooting of computational problems, model understanding, and model comparison. We review all these aspects of workflow in the context of several examples, keeping in mind that in practice we will be fitting many models for any given problem, even if only a subset of them will ultimately be relevant for our conclusions.

stat.ME↗

arXiv2020

Validating Bayesian Inference Algorithms with Simulation-Based Calibration

Verifying the correctness of Bayesian computation is challenging. This is especially true for complex models that are common in practice, as these require sophisticated model implementations and algorithms. In this paper we introduce \emph{simulation-based calibration} (SBC), a general procedure for validating inferences from Bayesian algorithms capable of generating posterior samples. This procedure not only identifies inaccurate computation and inconsistencies in model implementations but also provides graphical summaries that can indicate the nature of the problems that arise. We argue that SBC is a critical part of a robust Bayesian workflow, as well as being a useful tool for those developing computational algorithms and statistical software.

stat.ME↗

arXiv2020

Adaptive Path Sampling in Metastable Posterior Distributions

The normalizing constant plays an important role in Bayesian computation, and there is a large literature on methods for computing or approximating normalizing constants that cannot be evaluated in closed form. When the normalizing constant varies by orders of magnitude, methods based on importance sampling can require many rounds of tuning. We present an improved approach using adaptive path sampling, iteratively reducing gaps between the base and target. Using this adaptive strategy, we develop two metastable sampling schemes. They are automated in Stan and require little tuning. For a multimodal posterior density, we equip simulated tempering with a continuous temperature. For a funnel-shaped entropic barrier, we adaptively increase mass in bottleneck regions to form an implicit divide-and-conquer. Both approaches empirically perform better than existing methods for sampling from metastable distributions, including higher accuracy and computation efficiency.

stat.CO↗

arXiv2020

Bayesian hierarchical weighting adjustment and survey inference

We combine Bayesian prediction and weighted inference as a unified approach to survey inference. The general principles of Bayesian analysis imply that models for survey outcomes should be conditional on all variables that affect the probability of inclusion. We incorporate the weighting variables under the framework of multilevel regression and poststratification, as a byproduct generating model-based weights after smoothing. We investigate deep interactions and introduce structured prior distributions for smoothing and stability of estimates. The computation is done via Stan and implemented in the open source R package "rstanarm" ready for public use. Simulation studies illustrate that model-based prediction and weighting inference outperform classical weighting. We apply the proposal to the New York Longitudinal Study of Wellbeing. The new approach generates robust weights and increases efficiency for finite population inference, especially for subsets of the population.

stat.ME↗

arXiv2020

Accounting for Uncertainty During a Pandemic

We discuss several issues of statistical design, data collection, analysis, communication, and decision making that have arisen in recent and ongoing coronavirus studies, focusing on tools for assessment and propagation of uncertainty. This paper does not purport to be a comprehensive survey of the research literature; rather, we use examples to illustrate statistical points that we think are important.

stat.AP↗

arXiv2020

Improving multilevel regression and poststratification with structured priors

A central theme in the field of survey statistics is estimating population-level quantities through data coming from potentially non-representative samples of the population. Multilevel Regression and Poststratification (MRP), a model-based approach, is gaining traction against the traditional weighted approach for survey estimates. MRP estimates are susceptible to bias if there is an underlying structure that the methodology does not capture. This work aims to provide a new framework for specifying structured prior distributions that lead to bias reduction in MRP estimates. We use simulation studies to explore the benefit of these prior distributions and demonstrate their efficacy on non-representative US survey data. We show that structured prior distributions offer absolute bias reduction and variance reduction for posterior MRP estimates in a large variety of data regimes.

stat.ME↗

arXiv2020

Bayesian aggregation of average data: An application in drug development

Throughout the different phases of a drug development program, randomized trials are used to establish the tolerability, safety, and efficacy of a candidate drug. At each stage one aims to optimize the design of future studies by extrapolation from the available evidence at the time. This includes collected trial data and relevant external data. However, relevant external data are typically available as averages only, for example from trials on alternative treatments reported in the literature. Here we report on such an example from a drug development for wet age-related macular degeneration. This disease is the leading cause of severe vision loss in the elderly. While current treatment options are efficacious, they are also a substantial burden for the patient. Hence, new treatments are under development which need to be compared against existing treatments. The general statistical problem this leads to is meta-analysis, which addresses the question of how we can combine datasets collected under different conditions. Bayesian methods have long been used to achieve partial pooling. Here we consider the challenge when the model of interest is complex (hierarchical and nonlinear) and one dataset is given as raw data while the second dataset is given as averages only. In such a situation, common meta-analytic methods can only be applied when the model is sufficiently simple for analytic approaches. When the model is too complex, for example nonlinear, an analytic approach is not possible. We provide a Bayesian solution by using simulation to approximately reconstruct the likelihood of the external summary and allowing the parameters in the model to vary under the different conditions. We first evaluate our approach using fake-data simulations and then report results for the drug development program that motivated this research.

stat.AP↗

arXiv2020

Know your population and know your model: Using model-based regression and poststratification to generalize findings beyond the observed sample

Psychology research focuses on interactions, and this has deep implications for inference from non-representative samples. For the goal of estimating average treatment effects, we propose to fit a model allowing treatment to interact with background variables and then average over the distribution of these variables in the population. This can be seen as an extension of multilevel regression and poststratification (MRP), a method used in political science and other areas of survey research, where researchers wish to generalize from a sparse and possibly non-representative sample to the general population. In this paper, we discuss areas where this method can be used in the psychological sciences. We use our method to estimate the norming distribution for the Big Five Personality Scale using open source data. We argue that large open data sources like this and other collaborative data sources can be combined with MRP to help resolve current challenges of generalizability and replication in psychology.

stat.AP↗