Searcharxiv⌕ Search

arXiv subjects

Peter M. Aronow

Publications and source records attributed to Peter M. Aronow.

15 recordsLinked to original sources
arXiv2023

Inference in Spatial Experiments with Interference using the SpatialEffect Package

This paper presents methods for analyzing spatial experiments when complex spillovers, displacement effects, and other types of "interference" are present. We present a robust, design-based approach to analyzing effects in such settings. The design-based approach derives inferential properties for causal effect estimators from known features of the experimental design, in a manner analogous to inference in sample surveys. The methods presented here target a quantity of interest called the "average marginalized response," which is equal to the average effect of activating a treatment at an intervention node that is a given distance away, averaging ambient effects emanating from other intervention nodes. We provide a step-by-step tutorial based on the SpatialEffect package for R. We apply the methods to a randomized experiment on payments for community forest conservation in Uganda, showing how our methods reveal possibly substantial spatial spillovers that more conventional analyses cannot detect.

stat.ME↗
arXiv2021

Listwise Deletion in High Dimensions

We consider the properties of listwise deletion when both $n$ and the number of variables grow large. We show that when (i) all data has some idiosyncratic missingness and (ii) the number of variables grows superlogarithmically in $n$, then, for large $n$, listwise deletion will drop all rows with probability 1. Using two canonical datasets from the study of comparative politics and international relations, we provide numerical illustration that these problems may emerge in real world settings. These results suggest, in practice, using listwise deletion may mean using few of the variables available to the researcher.

stat.OT↗
arXiv2020

Spillover Effects in Experimental Data

We present current methods for estimating treatment effects and spillover effects under "interference", a term which covers a broad class of situations in which a unit's outcome depends not only on treatments received by that unit, but also on treatments received by other units. To the extent that units react to each other, interact, or otherwise transmit effects of treatments, valid inference requires that we account for such interference, which is a departure from the traditional assumption that units' outcomes are affected only by their own treatment assignment. Interference and associated spillovers may be a nuisance or they may be of substantive interest to the researcher. In this chapter, we focus on interference in the context of randomized experiments. We review methods for when interference happens in a general network setting. We then consider the special case where interference is contained within a hierarchical structure. Finally, we discuss the relationship between interference and contagion. We use the interference R package and simulated data to illustrate key points. We consider efficient designs that allow for estimation of the treatment and spillover effects and discuss recent empirical studies that try to capture such effects.

stat.AP↗
arXiv2019

Average treatment effects in the presence of unknown interference

We investigate large-sample properties of treatment effect estimators under unknown interference in randomized experiments. The inferential target is a generalization of the average treatment effect estimand that marginalizes over potential spillover effects. We show that estimators commonly used to estimate treatment effects under no interference are consistent for the generalized estimand for several common experimental designs under limited but otherwise arbitrary and unknown interference. The rates of convergence depend on the rate at which the amount of interference grows and the degree to which it aligns with dependencies in treatment assignment. Importantly for practitioners, the results imply that if one erroneously assumes that units do not interfere in a setting with limited, or even moderate, interference, standard estimators are nevertheless likely to be close to an average treatment effect if the sample is sufficiently large. Conventional confidence statements may, however, not be accurate.

math.ST↗
arXiv2019

A note on breaking ties among sample medians

Given samples $x_1,\cdots,x_n$, it is well known that any sample median value (not necessarily unique) minimizes the absolute loss $\sum_{i=1}^n |q-x_i|$. Interestingly, we show that the minimizer of the loss $\sum_{i=1}^n|q-x_i|^{1+ε}$ exhibits a singular perturbation behaviour that provides a unique definition for the sample median as $ε\rightarrow 0$. This definition is the unique point among all candidate median values that balances the $logarithmic$ moment of the empirical distribution. The result generalizes directly to breaking ties among sample quantiles when the quantile regression loss is modified in the same way.

stat.ME↗
arXiv2018

Estimating Average Causal Effects Under General Interference, with Application to a Social Network Experiment

This paper presents a randomization-based framework for estimating causal effects under interference between units, motivated by challenges that arise in analyzing experiments on social networks. The framework integrates three components: (i) an experimental design that defines the probability distribution of treatment assignments, (ii) a mapping that relates experimental treatment assignments to exposures received by units in the experiment, and (iii) estimands that make use of the experiment to answer questions of substantive interest. We develop the case of estimating average unit-level causal effects from a randomized experiment with interference of arbitrary but known form. The resulting estimators are based on inverse probability weighting. We provide randomization-based variance estimators that account for the complex clustering that can occur when interference is present. We also establish consistency and asymptotic normality under local dependence assumptions. We discuss refinements including covariate-adjusted effect estimators and ratio estimation. We evaluate empirical performance in realistic settings with a naturalistic simulation using social network data from American schools. We then present results from a field experiment on the spread of anti-conflict norms and behavior among school students.

math.ST↗
arXiv2016

A Note on "How Robust Standard Errors Expose Methodological Problems They Do Not Fix, and What to Do About It"

King and Roberts (2015, KR) claim that a disagreement between robust and classical standard errors exposes model misspecification. We emphasize that KR's claim only generally applies to parametric models: models that assume a restrictive form of the distribution of the outcome. Many common models in use in political science, including the linear model, are not necessarily parametric -- rather they may be semiparametric. Common estimators of model parameters such as ordinary least squares have both robust (corresponding to a semiparametric model) and classical (corresponding to a more restrictive model) standard error estimates. Given a properly specified semiparametric model and mild regularity conditions, the classical standard errors are not generally consistent, but the robust standard errors are. To illustrate this point, we consider the case of the regression estimate of a semiparametric linear model with no model misspecification, and show that robust standard errors may nevertheless systematically differ from classical standard errors. We show that a disagreement between robust and classical standard errors is not generally suitable as a diagnostic for regression estimators, and that KR's reanalyses of Neumayer (2003) and Büthe and Milner (2008) are predicated on strong assumptions that the original authors did not invoke nor require.

stat.ME↗
arXiv2016

Local average causal effects and superefficiency

Recent approaches in causal inference have proposed estimating average causal effects that are local to some subpopulation, often for reasons of efficiency. These inferential targets are sometimes data-adaptive, in that they are dependent on the empirical distribution of the data. In this short note, we show that if researchers are willing to adapt the inferential target on the basis of efficiency, then extraordinary gains in precision can be obtained. Specifically, when causal effects are heterogeneous, any asymptotically normal and root-$n$ consistent estimator of the population average causal effect is superefficient for a data-adaptive local average causal effect. Our result illustrates the fundamental gain in statistical certainty afforded by indifference about the inferential target.

math.ST↗
arXiv2016

Confidence intervals for means under constrained dependence

We develop a general framework for conducting inference on the mean of dependent random variables given constraints on their dependency graph. We establish the consistency of an oracle variance estimator of the mean when the dependency graph is known, along with an associated central limit theorem. We derive an integer linear program for finding an upper bound for the estimated variance when the graph is unknown, but topological and degree-based constraints are available. We develop alternative bounds, including a closed-form bound, under an additional homoskedasticity assumption. We establish a basis for Wald-type confidence intervals for the mean that are guaranteed to have asymptotically conservative coverage. We apply the approach to inference from a social network link-tracing study and provide statistical software implementing the approach.

math.ST↗
arXiv2015

Identification of homophily and preferential recruitment in respondent-driven sampling

Respondent-driven sampling (RDS) is a link-tracing procedure for surveying hidden or hard-to-reach populations in which subjects recruit other subjects via their social network. There is significant research interest in detecting clustering or dependence of epidemiological traits in networks, but researchers disagree about whether data from RDS studies can reveal it. Two distinct mechanisms account for dependence in traits of recruiters and recruitees in an RDS study: homophily, the tendency for individuals to share social ties with others exhibiting similar characteristics, and preferential recruitment, in which recruiters do not recruit uniformly at random from their available alters. The different effects of network homophily and preferential recruitment in RDS studies have been a source of confusion in methodological research on RDS, and in empirical studies of the social context of health risk in hidden populations. In this paper, we give rigorous definitions of homophily and preferential recruitment and show that neither can be measured precisely in general RDS studies. We derive nonparametric identification regions for homophily and preferential recruitment and show that these parameters are not point identified unless the network takes a degenerate form. The results indicate that claims of homophily or recruitment bias measured from empirical RDS studies may not be credible. We apply our identification results to a study involving both a network census and RDS on a population of injection drug users in Hartford, CT.

stat.AP↗
arXiv2015

Cluster-Robust Variance Estimation for Dyadic Data

Dyadic data are common in the social sciences, although inference for such settings involves accounting for a complex clustering structure. Many analyses in the social sciences fail to account for the fact that multiple dyads share a member, and that errors are thus likely correlated across these dyads. We propose a nonparametric sandwich-type robust variance estimator for linear regression to account for such clustering in dyadic data. We enumerate conditions for estimator consistency. We also extend our results to repeated and weighted observations, including directed dyads and longitudinal data, and provide an implementation for generalized linear models such as logistic regression. We examine empirical performance with simulations and applications to international relations and speed dating.

stat.ME↗
arXiv2015

Nonparametric Identification for Respondent-Driven Sampling

Respondent-driven sampling is a survey method for hidden or hard-to-reach populations in which sampled individuals recruit others in the study population via their social links. The most popular estimator for for the population mean assumes that individual sampling probabilities are proportional to each subject's reported degree in a social network connecting members of the hidden population. However, it remains unclear under what circumstances these estimators are valid, and what assumptions are formally required to identify population quantities. In this short note we detail nonparametric identification results for the population mean when the sampling probability is assumed to be a function of network degree known to scale. Importantly, we establish general conditions for the consistency of the popular Volz-Heckathorn (VH) estimator. Our results imply that the conditions for consistency of the VH estimator are far less stringent than those suggested by recent work on diagnostics for RDS. In particular, our results do not require random sampling or the existence of a network connecting the population.

stat.ME↗
arXiv2014

Combining List Experiment and Direct Question Estimates of Sensitive Behavior Prevalence

Survey respondents may give untruthful answers to sensitive questions when asked directly. In recent years, researchers have turned to the list experiment (also known as the item count technique) to overcome this difficulty. While list experiments may be less prone to bias than direct questioning, list experiments are also more susceptible to sampling variability. We show that researchers do not have to abandon direct questioning altogether in order to gain the advantages of list experimentation. We develop a nonparametric estimator of the prevalence of sensitive behaviors that combines list experimentation and direct questioning. We prove that this estimator is asymptotically more efficient than the standard difference-in-means estimator, and we provide a basis for inference using Wald-type confidence intervals. Additionally, leveraging information from the direct questioning, we derive two nonparametric placebo tests of the identifying assumptions for the list experiment. We demonstrate the effectiveness of our combined estimator and placebo tests with an original survey experiment.

stat.AP↗
arXiv2014

Sharp bounds on the variance in randomized experiments

We propose a consistent estimator of sharp bounds on the variance of the difference-in-means estimator in completely randomized experiments. Generalizing Robins [Stat. Med. 7 (1988) 773-785], our results resolve a well-known identification problem in causal inference posed by Neyman [Statist. Sci. 5 (1990) 465-472. Reprint of the original 1923 paper]. A practical implication of our results is that the upper bound estimator facilitates the asymptotically narrowest conservative Wald-type confidence intervals, with applications in randomized controlled and clinical trials.

math.ST↗