Do School Districts Affect NYC House Prices? Identifying Border Differences Using a Bayesian Nonparametric Approach to Geographic Regression Discontinuity Designs

Rischard, Maxime; Branson, Zach; Miratrix, Luke; Bornn, Luke

doi:10.1080/01621459.2020.1817749

Cited by 6 publications

(9 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…which is a point process extension to the estimands seen in previous spatial regression discontinuity designs Rischard et al, 2020).…”

Section: Single Border Estimandsmentioning

confidence: 99%

“…Important aspects specific to the geographic design are highlighted in . This design has been used to estimate the effect of private police departments on crime (MacDonald et al, 2016), the impact of voter initiatives on voter turnout , the effect of the Civil Rights Act of 1875 (Harvey, 2020), and whether school districts impact housing prices (Rischard et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Robust inference for geographic regression discontinuity designs: assessing the impact of police precincts

Kendall¹,

Beck²,

Antonelli³

2021

Preprint

View full text Add to dashboard Cite

We study variation in policing outcomes attributable to differential policing practices in New York City (NYC) using geographic regression discontinuity designs. By focusing on small geographic windows near police precinct boundaries we can estimate local average treatment effects of precincts on arrest rates. The standard geographic regression discontinuity design relies on continuity assumptions of the potential outcome surface or a local randomization assumption within a window around the boundary. While these assumptions are often thought to be more realistic than other assumptions used to infer causality from observational data, they can easily be violated in realistic applications. We develop a novel and robust approach to testing whether there are differences in policing outcomes that are caused by differences in police precincts across NYC. In particular, our test is robust to violations of the assumptions traditionally made in geographic regression discontinuity designs and is valid under much weaker assumptions. We use a unique form of resampling to identify new geographic boundaries that are known to have no treatment effect, which provides a valid estimate of our estimator's null distribution even under violations of standard assumptions. We find that this procedure gives substantially different results in the analysis of NYC arrest rates than those that rely on standard assumptions, thereby providing more robust estimates of the nature of the effect of police precincts on arrest rates in NYC.

show abstract

“…which is a point process extension to the estimands seen in previous spatial regression discontinuity designs Rischard et al, 2020).…”

Section: Single Border Estimandsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Robust inference for geographic regression discontinuity designs: assessing the impact of police precincts

Kendall¹,

Beck²,

Antonelli³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Alternatively, one can use kernel methods that compute a smoothly weighted average of the data points to create an interpolated regression that does not depend on a specific parametric form [ 20 ]. Other studies have considerd using Gaussian process regression for regression discontinuity as well [ 10 , 26 ]. Here, instead of fitting a parametric form such as linear regression, the regression is modelled by a GP, which results in a flexible, nonparametric model and more accurate effect size estimates compared to when using linear regression.…”

Section: Related Workmentioning

confidence: 99%

“…The idea behind these approaches is that, around the assignment threshold, observations are distributed essentially randomly, so that locally the conditions of RCT are recreated [ 8 , 9 ]. The methodological pipeline of quasi-experimental designs like these generally consists of three steps [ 10 ]. First, a regression (typically linear) is fit to each of the two groups individually.…”

Section: Introductionmentioning

confidence: 99%

Bayesian model averaging for nonparametric discontinuity design

et al. 2022

View full text Add to dashboard Cite

Quasi-experimental research designs, such as regression discontinuity and interrupted time series, allow for causal inference in the absence of a randomized controlled trial, at the cost of additional assumptions. In this paper, we provide a framework for discontinuity-based designs using Bayesian model averaging and Gaussian process regression, which we refer to as ‘Bayesian nonparametric discontinuity design’, or BNDD for short. BNDD addresses the two major shortcomings in most implementations of such designs: overconfidence due to implicit conditioning on the alleged effect, and model misspecification due to reliance on overly simplistic regression models. With the appropriate Gaussian process covariance function, our approach can detect discontinuities of any order, and in spectral features. We demonstrate the usage of BNDD in simulations, and apply the framework to determine the effect of running for political positions on longevity, of the effect of an alleged historical phantom border in the Netherlands on Dutch voting behaviour, and of Kundalini Yoga meditation on heart rate.

show abstract

“…This is important, as recent work has found that most existing methods that are used for RDD designs tend to understate uncertainty and provide anti-conservative inference (Stommes et al, 2021). This work has been further extended to spatial RDD settings where the goal is estimation of the treatment effect curve on a two-dimensional boundary (Rischard et al, 2020). Gaussian processes are also perfectly suited to this setting, as they can immediately handle a bivariate running variable X i = (X i1 , X i2 ), such as latitude and longitude.…”

Section: Regression Discontinuity Designsmentioning

confidence: 99%

The How and Why of Bayesian Nonparametric Causal Inference

Linero¹,

Antonelli²

2021

Preprint

View full text Add to dashboard Cite

Spurred on by recent successes in causal inference competitions, Bayesian nonparametric (and high-dimensional) methods have recently seen increased attention in the causal inference literature. In this paper, we present a comprehensive overview of Bayesian nonparametric applications to causal inference. Our aims are to (i) introduce the fundamental Bayesian nonparametric toolkit; (ii) discuss how to determine which tool is most appropriate for a given problem; and (iii) show how to avoid common pitfalls in applying Bayesian nonparametric methods in high-dimensional settings. Unlike standard fixed-dimensional parametric problems, where outcome modeling alone can sometimes be effective, we argue that most of the time it is necessary to model both the selection and outcome processes.

show abstract

Do School Districts Affect NYC House Prices? Identifying Border Differences Using a Bayesian Nonparametric Approach to Geographic Regression Discontinuity Designs

Cited by 6 publications

References 33 publications

Robust inference for geographic regression discontinuity designs: assessing the impact of police precincts

Robust inference for geographic regression discontinuity designs: assessing the impact of police precincts

Bayesian model averaging for nonparametric discontinuity design

The How and Why of Bayesian Nonparametric Causal Inference

Contact Info

Product

Resources

About