Instrumental Variable Quantile Regression Research Articles

Association of ICU Admission and Outcomes in Sepsis and Acute Respiratory Failure.

Rationale: Many decisions to admit patients to the ICU are not grounded in evidence regarding who benefits from such triage, straining ICU capacity and limiting its cost-effectiveness. Objectives: To measure the benefits of ICU admission for patients with sepsis or acute respiratory failure. Methods: At 27 United States hospitals across two health systems from 2013 to 2018, we performed a retrospective cohort study using two-stage instrumental variable quantile regression with a strong instrument (hospital capacity strain) governing ICU versus ward admission among high-acuity patients (i.e., laboratory-based acute physiology score v2 ⩾ 100) with sepsis and/or acute respiratory failure who did not require mechanical ventilation or vasopressors in the emergency department. Measurements and Main Results: Among patients with sepsis (n = 90,150), admission to the ICU was associated with a 1.32-day longer hospital length of stay (95% confidence interval [CI], 1.01-1.63; P < 0.001) (when treating deaths as equivalent to long lengths of stay) and higher in-hospital mortality (odds ratio, 1.48; 95% CI, 1.13-1.88; P = 0.004). Among patients with respiratory failure (n = 45,339), admission to the ICU was associated with a 0.82-day shorter hospital length of stay (95% CI, -1.17 to -0.46; P < 0.001) and reduced in-hospital mortality (odds ratio, 0.75; 95% CI, 0.57-0.96; P = 0.04). In sensitivity analyses of length of stay, excluding, ignoring, or censoring death, results were similar in sepsis but not in respiratory failure. In subgroup analyses, harms of ICU admission for patients with sepsis were concentrated among older patients and those with fewer comorbidities, and the benefits of ICU admission for patients with respiratory failure were concentrated among older patients, highest-acuity patients, and those with more comorbidities. Conclusions: Among high-acuity patients with sepsis who did not require life support in the emergency department, initial admission to the ward, compared with the ICU, was associated with shorter length of stay and improved survival, whereas among patients with acute respiratory failure, triage to the ICU compared with the ward was associated with improved survival.

Poverty and terrorism in Africa: Understanding the nexus based on existing levels of terrorism

AbstractPrevailing literature recognizes that poverty may not act as a significant factor that spurs the level of terrorist activities. However, literature on the relationship between poverty and terrorism in Africa remains sparse. This study, therefore, investigates the influence of poverty on terrorism in 26 African countries from 2000 to 2017 in a panel data environment. Findings from the two‐stage least square regression, with standard errors which are heteroskedastic and autocorrelation consistent, show that poverty significantly raises the level of terrorism incidents. However, after accounting for unobserved heterogeneity, we find that the positive relationship between poverty and terrorism becomes insignificant. Further findings employing the instrumental variable quantile regression with fixed effects show that the existing levels of terrorism incidents matter for the impact of poverty on terrorism incidents. The result shows that the impact of poverty on terrorism is only significant at the median. Furthermore, economic growth, political stability, income inequality, and unemployment matter for terrorism in Africa. The study highlights that poverty is not a determining factor for terrorism in Africa when compared to economic growth, political stability, and unemployment. Recommendations are such that inclusive growth, political stability, and unemployment reduction should be prioritized in Africa.

Moment estimation for censored quantile regression

In influential articles Powell (Journal of Econometrics 25(3):303–325, 1984; Journal of Econometrics 32(1):143–155, 1986) proposed optimization-based censored least absolute deviations estimator (CLAD) and general censored quantile regression estimator (CQR). It has been recognized, however, that this optimization-based estimator may perform poorly in finite samples (e.g., Khan and Powell, Journal of Econometrics 103(1–2):73–110, 2001; Fitzenberger, Handbook of Statistics. Elsevier, 1996; Fitzenberger and Winker, Computational Statistics & Data Analysis 52(1):88–108, 2007; Koenker, Journal of Statistical Software 27(6), 2008). In this paper we propose a moment-based censored quantile regression estimator (MCQR). While both the CQR and MCQR estimators have the same large sample properties, our simulation results suggest certain advantage of our moment-based estimator (MCQR). In addition, the empirical likelihood methods for the uncensored model (e.g., Whang 2006; Otsu, Journal of Econometrics 142(1):508–538, 2008) can readily be adapted to the censored model within our method of moment estimation framework. When both censoring and endogeneity are present, we develop an instrumental variable censored quantile regression estimator (IVCQR) by combining the insights of Chernozhukov and Hansen’s (Journal of Econometrics 132(2):491–525, 2006; Journal of Econometrics 142(1):379–398, 2008) instrumental variables quantile regression estimator (IVQR) and the MCQR. Simulation results indicate that the IVCQR estimator performs well.

Heterogeneous impacts of information technology adoption on pesticide and fertiliser expenditures: Evidence from wheat farmers in China*

This study examines the impacts of mobile information technology adoption on agrochemical expenditures, using data collected from 551 wheat farmers in China. Unlike previous studies that analyse a homogeneous relationship between the adoption of information technologies and farm input use, in this study, an instrumental variable quantile regression approach is utilised to capture the heterogeneous impacts of smartphone use on pesticide and fertiliser expenditures. Findings reveal that smartphone use affects pesticide and fertiliser expenditures heterogeneously, and its impacts on pesticide expenditure are larger than those on fertiliser expenditure. Specifically, at the lowest 20th quantile, smartphone use significantly increases pesticide expenditure by 33 per cent and fertiliser expenditure by 18 per cent. However, at the higher 60th and 80th quantiles, smartphone use significantly decreases pesticide expenditure by 36–39 per cent and fertiliser expenditure by 14–19 per cent. Our findings suggest that guiding farmers’ agrochemical usage behaviours through smartphone‐based information intervention can be a practical strategy to help reduce the excessive usage of chemical pesticides and fertilisers and preserve the environment and human health.

Who benefits more? The heterogeneous impact of highways on employment growth

ABSTRACT This paper explores the relationship between the stock of interstate highways on employment growth of counties in Texas between 1983 and 2012. Using the Chernozhukov-Hansen instrumental variables quantile regression (IVQR) method, we examine the heterogeneous impact that highways have on employment growth at different quantiles of the conditional distribution while at the same time controlling for potential endogeneity. The results show that the employment growth effect monotonically increases as one shifts from the lower tail to the upper tail of the distribution. The range is 0.15 to 0.44, with the highest at the 95th quantile, compared to 0.189 for OLS and 0.213 for 2SLS. A 10 percent increase in interstate highway kilometers in 1983 led to about a 1.5 to 4.4 percentage point increase in county employment over a 29-year period. Our results also indicate that the counties with low initial levels of employment grew faster than those with a high initial level of employment and that this convergence monotonically decreases from the lower tail to the upper tail of the growth distribution.

Penalized quantile regression for spatial panel data with fixed effects

This paper studies a penalized quantile regression for spatial panel model with fixed effects, where a penalized method is used to control additional variability by shrinking the individual fixed effects to a common value with a tuning parameter. The bias is reduced by employing the instrumental variables quantile regression method. Limiting properties of the proposed estimator are derived, say consistency and asymptotic normality. Monte Carlo simulations results are provided to illustrate the finite sample behavior of the proposed estimation methods. The use of the procedures is illustrated through an empirical application to analyze a spatial panel data related to military expenditure from 144 countries over 15 years.

Debiased/Double Machine Learning for Instrumental Variable Quantile Regressions

In this study, we investigate the estimation and inference on a low-dimensional causal parameter in the presence of high-dimensional controls in an instrumental variable quantile regression. Our proposed econometric procedure builds on the Neyman-type orthogonal moment conditions of a previous study (Chernozhukov et al. 2018) and is thus relatively insensitive to the estimation of the nuisance parameters. The Monte Carlo experiments show that the estimator copes well with high-dimensional controls. We also apply the procedure to empirically reinvestigate the quantile treatment effect of 401(k) participation on accumulated wealth.

Does Mobile Internet Use Affect the Subjective Well-being of Older Chinese Adults? An Instrumental Variable Quantile Analysis

Although mobile Internet has become an integral part of people’s daily life in the world today, research on whether and to what extent mobile Internet use (MIU) affects the subjective well-being (SWB) of older adults is scant. This paper focuses on the SWB-MIU gradients among older Chinese adults and contributes to the literature by exploring the relationship between MIU and SWB across the entire distribution of older adults’ SWB. Using nationally representative data from the China Family Panel Studies, we find that MIU has a significant positive impact on the SWB of older adults in China. Employing an instrumental variable quantile regression approach, we provide robust evidence of an asymmetric distributional impact of MIU on SWB among older Chinese adults. In particular, we find that individuals with lower levels of SWB benefit more from the usage of mobile Internet. The results further indicate that MIU can improve the relationship between older adults and their adult children, and is conducive to the maintenance of kin ties, which we interpret as potential mechanisms through which MIU positively affects SWB among older Chinese adults.

What are the shocks of climate change on clean energy investment: A diversified exploration

Abstract Climate change may affect energy consumption and thus bring shocks to clean energy investment. By employing instrumental variable quantile regression model, this research investigates the effects of climate change on clean energy investment according to global panel data spanning a long period. In addition to the sample of all countries, we also carry out investigations of sub-samples for OECD and non-OECD countries to analyze whether there exists heterogeneity in development levels. Overall, we provide a diversified exploration on how climate change shocks impact clean energy investment. More specifically, we find that effects of climate change on clean energy investment vary significantly in countries with different levels of clean energy investment; i.e., climate change is likely to promote clean energy investment in countries whose clean investments are greater, and there exists significant heterogeneity in the subsamples. From these results, we offer policy implications such as carrying out differentiated clean energy plans based on the specific environment, implementing policies on clean energy, and copying the practices in some OECD countries that have high levels of clean energy investment.

Competition, price dispersion and capacity constraints: the case of the U.S. corn seed industry

abstract In this paper we examine the effect of competition on price dispersion and argue that the effect is contingent on the ability of firms to meet market demand. Our comparative static results show that competition among symmetrically capacity-unconstrained firms, or among firms with asymmetric capacities leads to an overall price increase along the distribution function. To investigate these findings empirically, we use a novel data set from the U.S. corn seed industry with firm and farm level sales information for conventional and genetically modified corn seeds between 2004 and 2009. We estimate the empirical model using the Fixed Effect Instrumental Variable Quantile Regression and find evidence consistent with the theory. The analysis also shows that capacity-unconstrained seed firms charge a price premium, confirming the positive relationship between product availability and pricing found in our theoretical model.

Decentralization estimators for instrumental variable quantile regression models

The instrumental variable quantile regression (IVQR) model (Chernozhukov and Hansen (2005)) is a popular tool for estimating causal quantile effects with endogenous covariates. However, estimation is complicated by the nonsmoothness and nonconvexity of the IVQR GMM objective function. This paper shows that the IVQR estimation problem can be decomposed into a set of conventional quantile regression subproblems which are convex and can be solved efficiently. This reformulation leads to new identification results and to fast, easy to implement, and tuning‐free estimators that do not require the availability of high‐level “black box” optimization routines.

Does Building Highways Reduce Traffic Congestion?

In a seminal study, Duranton and Turner (2011) finds evidence that points to the existence of the fundamental law of highway congestion in the US. They build a causal model using an instrumental variable (IV) approach that yields an estimate of 1.03 for the elasticity of vehicle miles traveled (VMT) to the stock of interstate highways in US metropolitan areas. The result means that government efforts to alleviate traffic congestion by expanding capacity are likely to fail — any increase in the stock of highways is accompanied by a commensurate increase in VMT, leaving travel times unaffected. In this article, we explore the impact of unobserved heterogeneity on the fundamental law. We begin by using a simple partial equilibrium model to demonstrate how metropolitan statistical areas (MSAs) that are identical in most respects but have different initial congestion levels respond differently to added capacity due to individual differences. These differences in MSAs gives rise to heterogeneity in the elasticity of VMT to capacity. We derive conditions under which the elasticity decreases with the initial congestion level. We then revisit the empirical analysis in Duranton and Turner (2011) using the instrumental variable quantile regression (IV-QR) model due to Chernozhukov and Hansen (2005, 2006, 2008). The IV-QR model allows us to incorporate variation in the elasticity due to the presence of unobserved differences across MSAs. Moreover, it allows us to evaluate the impact of changes in the stock of interstate highways on the entire conditional distribution of VMT, not just the impact on the conditional mean as in Duranton and Turner (2011). The IV-QR estimates show that as predicted by the simple partial equilibrium model, the elasticity declines as one goes up the quantile ladder, being more than one at the lower quantiles and less than one at the higher quantiles. The median IV-QR estimate being close to one. The IV-QR model implies that among observationally identical cities, expanding road capacity can lower the number of cities experiencing severe congestion, although the mean or median congestion levels are likely to remain constant. We also estimate the impact of increased road capacity on the unconditional distribution of VMT using the generalized quantile regression (GQR) model due to Powell (Forthcoming). The GQR estimates mirror the IV-QR estimates, but their conclusions are starker at the upper quantiles: building highways have no statistically significant impact on VMT at the highest quantiles. The GQR results imply that building roads can lower the total number of cities having different observed characteristics experiencing severe congestion levels. We further explore the mechanisms that drive the empirical findings by running simulations using a spatial general equilibrium model with an extensive road network calibrated to the Greater Los Angeles (LA) Region. In the general equilibrium model, besides commute trips, consumers have to travel to different zones to acquire consumption goods. The model considers route choice in the network and mode choice. Building roads affect road traffic through three channels: the total amount of consumption, mode choice, and the substitutions among goods sold at different locations. We find that the elasticity of VMT to capacity in LA is 0.321, and the elasticity decreases consistently with the initial congestion level. We report the welfare effect and the changes in other travel-related variables. Our results have important policy implications in that they show that while building roads is unlikely to change the mean or median congestion level, it can reduce the number of cities experiencing high congestion levels.

Dynamic Spatial Network Quantile Autoregression

We propose a dynamic network quantile regression model to investigate the quantile connectedness using a predetermined network information. We extend the existing network quantile autoregression model of Zhu et al. (2019b) by explicitly allowing the contemporaneous network effects and controlling for the common factors across quantiles. To cope with the endogeneity issue due to simultaneous network spillovers, we adopt the instrumental variable quantile regression (IVQR) estimation and derive the consistency and asymptotic normality of the IVQR estimator using the near epoch dependence property of the network process. Via Monte Carlo simulations, we confirm the satisfactory performance of the IVQR estimator across different quantiles under the different network structures. Finally, we demonstrate the usefulness of our proposed approach with an application to the dataset on the stocks traded in NYSE and NASDAQ in 2016.

Exploring Water Consumption in Dhaka City Using Instrumental Variables Regression Approaches

This paper explores water consumption in Dhaka city for better understanding of its usage, and considers the implications of findings from distributive rationale. Using 459 household survey data collected by BRAC Institute of Governance and Development (BIGD), this study estimates income elasticities of water consumption after controlling the effects of other covariates including wealth-proxies, location, household size, water bill and spatial zones using the instrumental variable regression (IVREG) and instrumental variable quantile regression (IVQREG) approaches. The latter has an additive advantage over the former as the IVQREG provides a more accurate picture of the relationship of water consumption with the income throughout the entire water consumption distribution. Using the fixed pay variable as instrument, findings reveal the strong evidence that income is endogenous. The IVQREG results show that income elasticities are heterogeneous and vary significantly across the water quantiles, implying inequality in water consumption. It also provides strong systematic evidence as income elasticity of water consumption decreases with the increase in percentile. Significant spatial inequality in water consumption from IVREG approach disappears as we use IVQREG. This also strongly supports the systematic evidence obtained. Therefore, it is imperative to introduce different tariff structures among different water consumer groups for bringing equity in water consumption and revenue generation. However, Dhaka Water Supply & Sewerage Authority (DWASA) must ensure smart water meter before implementing such tariff structure as we face severe challenges while measuring residential water consumption.

The effects of international remittances on expenditure patterns of the left‐behind households in Sub‐Saharan Africa

AbstractThis article explores the effects of international remittances on the expenditure patterns of households in Sub‐Saharan Africa (SSA). This article focuses on five countries in SSA, which are some of the destinations that account for the highest receipt of international remittances. We analyze both aggregate and distributional effects of international remittances on expenditure patterns of households. To investigate the distributional effect of international remittances, we adopt the instrumental variable quantile (IV‐quantile) regression framework that allows us to simultaneously address the endogeneity of international remittances and possible heterogeneity in the impact of international remittances on households’ expenditure patterns. We instrument for international remittances by using the economic conditions in migrants’ countries as an instrument for international remittances. Our results show that the receipt of international remittances increases expenditures on food, durables, education, and health. Using the IV‐quantile regression, we find the effects of international remittances on household expenditure on food, durables, education, and health increase across the different expenditure quantiles.

Hollowing out the middle? Remittances, poverty, and income inequality in Nigeria

ABSTRACT This paper investigates the impact of remittances on poverty and inequality in Nigeria. In contrast to the existing literature, our methodology of instrumental variable quantile regression (IVQR) explicitly demonstrates the differential marginal impact of remittances for households at different levels of the conditional expenditure distribution. In tracing this heterogeneous impact, we are further able to address the effect of remittances on poverty and inequality simultaneously in one econometric model. Our results based on the Nigerian Migration Household Survey 2009 show that remittances reduce poverty by increasing household expenditures reveal a positive marginal impact of remittances at all but the highest quantiles of the conditional distribution of household expenditure, with the impact being the greatest up to the 12th quantile. While this unambiguously supports the poverty alleviation role of remittances documented in the literature, the distributional impact is more nuanced: The marginal effect of remittances follows a U-shape over most of the household expenditure distribution, which suggests that remittances may ‘hollow out’ the middle class. Specifically, households lying between the 13th to the 35th quantile gain less from receiving remittances than households on either side of this range.

The effect of board gender diversity on corporate social performance: An instrumental variable quantile regression approach

Investigating a sample of firms from the S&P 500 Index (between 2004 and 2015), the paper addresses the relationship between the representation of Women on Corporate Boards (WOCB) and the Corporate Social Performance (CSP). It adopts the instrumental variable quantile regression panel data (IV-QRPD) model, suggested by Powell (2016), to solve endogeneity and heterogeneity issues. The article shows a strong “threshold effect”: the impact of board feminization on CSP is nonlinear and changes along the quantiles of the performance distribution. Highlighting a contextual and multilevel phenomenon, the paper contributes to the CG literature offering main achievements on board's dynamics and inclusiveness.

Intergenerational transmission of economic success in Austria with a focus on migration and gender

In this paper, we analyse the intergenerational transmission of economic success in Austria using the European Union Statistics on Income and Living Conditions 2011 dataset (EU-SILC 2011). Starting point of the investigation is a two-step estimation procedure, where we detect a significant positve intergenerational association between the economic situation of the parental household and educational attainments as well as gross hourly wages of their male and female descendants independently. Furthermore, we shed some light on the intergenerational social mobility black box by explaining the direct effect of the income situation of the parental generation on attainable wages of the child generation within the transmission system. It turns out that this effect is significantly underestimated when applying ordinary least squares regressions only, where the perceived socioeconomic status is taken exogenous. Finally, we apply instrumental variable quantile regressions to demonstrate that the direct intergenerational economic association between parents and their descendants is strongest for top earners. Alongside this proceeding, we introduce an alternative way to think about the impact of a bundle of additional intergenerational transmission channels like cognitive ability, noncognitive personal traits, and aspects of physical appearance on an empirical level. Overall, the findings of the paper, where special attention is paid to migrants, offer a better understanding on the intergenerational transmission of economic success mechanism and to what extent this process influences income persistence between generations.

INSTRUMENTAL VARIABLE QUANTILE REGRESSION WITH MISCLASSIFICATION

AbstractThis article investigates the instrumental variable quantile regression model (Chernozhukov and Hansen, 2005,Econometrica73, 245–261; 2013,Annual Review of Economics, 5, 57–81) with a binary endogenous treatment. It offers two identification results when the treatment status is not directly observed. The first result is that, remarkably, the reduced-form quantile regression of the outcome variable on the instrumental variable provides a lower bound on the structural quantile treatment effect under the stochastic monotonicity condition. This result is relevant, not only when the treatment variable is subject to misclassification, but also when any measurement of the treatment variable is not available. The second result is for the structural quantile function when the treatment status is measured with error; the sharp identified set is characterized by a set of moment conditions under widely used assumptions on the measurement error. Furthermore, an inference method is provided in the presence of other covariates.

Explaining local governments' cost efficiency: Controllable and uncontrollable factors

Efficient and effective management of public resources is essential at all levels of government. This issue has gained momentum due to the strains that affected public sector finances after the onset of the 2007/08 crisis in many countries, particularly in Europe. In this article, we evaluate the influence of environmental variables that affect local government efficiency in one European country, Spain, during the crisis years (2009–2015). To this end, and considering the possible influence of both controllable and uncontrollable factors, we use an approach that is able to analyse their impact across the conditional distribution of performance, and which controls for the (likely) existence of endogeneity among regressors. Results show an asymmetry in the determinants of performance since the role played by some variables depends largely on the conditional distribution of efficiency. Results are robust not only because of the instrumental variables quantile regression approach considered, but also due to the variety of models specified to measure municipalities' performance as well as the diversity of covariates.