Auto IV: Counterfactual Prediction via Automatic Instrumental Variable Decomposition

Instrumental variables (IVs), sources of treatment randomization that are conditionally independent of the outcome, play an important role in causal inference with unobserved confounders. However, the existing IV-based counterfactual prediction methods need well-predefined IVs, while it’s an art rather than science to find valid IVs in many real-world scenes. Moreover, the predefined hand-made IVs could be weak or erroneous by violating the conditions of valid IVs. These thorny facts hinder the application of the IV-based counterfactual prediction methods. In this article, we propose a novel Automatic Instrumental Variable decomposition (AutoIV) algorithm to automatically generate representations serving the role of IVs from observed variables (IV candidates). Specifically, we let the learned IV representations satisfy the relevance condition with the treatment and exclusion condition with the outcome via mutual information maximization and minimization constraints, respectively. We also learn confounder representations by encouraging them to be relevant to both the treatment and the outcome. The IV and confounder representations compete for the information with their constraints in an adversarial game, which allows us to get valid IV representations for IV-based counterfactual prediction. Extensive experiments demonstrate that our method generates valid IV representations for accurate IV-based counterfactual prediction.

A Generalized Non-Parametric Instrumental Variable-Control Function Approach to Estimation in Nonlinear Settings

When the endogenous variables enter non-parametrically into the regression equation standard linear instrumental variables approaches fail. Two existing solutions are the non-parametric instrumental variables (NPIVs) estimators, which are based on a set of conditional moment restrictions (CMRs), and the control function (CF) estimators, which use conditional mean independence (CMI) restrictions. Our first contribution is to show that – similar to CMI – the CMR place shape restrictions on the conditional expectation of the error given the instruments and endogenous variables that are sufficient for identification, and we call our new estimator based on these restrictions the CMR-CF estimator. Our second contribution is to develop an estimator for non-linear and non-parametric settings that can combine both CMR and CMI restrictions, which cannot be done in either the NPIV nor the non-parametric CF setting. This new “Generalized CMR-CF” uses both CMR and CMI restrictions together by allowing the conditional expectation of the structural error to depend on both instruments and control variables. When sieves are used to approximate both the structural function and the CF our estimator reduces to a series of least squares regressions. Our Monte Carlos illustrate that our new estimator performs well across several economic settings.

Novel insights into the consequences of obesity: a phenotype-wide Mendelian randomization study

Obesity is thought to significantly impact the quality of life. In this study, we sought to evaluate the health consequences of obesity on the risk of a broad spectrum of human diseases. The causal effects of exposing to obesity on health outcomes were inferred using Mendelian randomization (MR) analyses using a fixed effects inverse-variance weighted model. The instrumental variables were SNPs associated with obesity as measured by body mass index (BMI) reported by GIANT consortium. The spectrum of outcome consisted of the phenotypes from published GWAS and the UK Biobank. The MR-Egger intercept test was applied to estimate horizontal pleiotropic effects, along with Cochran’s Q test to assess heterogeneity among the causal effects of instrumental variables. Our MR results confirmed many putative disease risks due to obesity, such as diabetes, dyslipidemia, sleep disorder, gout, smoking behaviors, arthritis, myocardial infarction, and diabetes-related eye disease. The novel findings indicated that elevated red blood cell count was inferred as a mediator of BMI-induced type 2 diabetes in our bidirectional MR analysis. Intriguingly, the effects that higher BMI could decrease the risk of both skin and prostate cancers, reduce calorie intake, and increase the portion size warrant further studies. Our results shed light on a novel mechanism of the disease-causing roles of obesity.

GIVE Statistic for Goodness of Fit in Instrumental Variables Models with Application to COVID Data

Since COVID-19 outbreak, scientists have been interested to know whether there is any impact of the Bacillus Calmette-Guerin (BCG) vaccine against COVID-19 mortality or not. It becomes more relevant as a large population in the world may have latent tuberculosis infection (LTBI), for which a person may not have active tuberculosis but persistent immune responses stimulated by Mycobacterium tuberculosis antigens, and that means, both LTBI and BCG generate immunity against COVID-19. In order to understand the relationship between LTBI and COVID-19 mortality, this article proposes a measure of goodness of fit, viz., Goodness of Instrumental Variable Estimates (GIVE) statistic, of a model obtained by Instrumental Variables estimation. The GIVE helps in finding the appropriate choice of instruments, which provides a better fitted model. In the course of study, the large sample properties of the GIVE statistic are investigated. As indicated before, the COVID-19 data is analysed using the GIVE statistic, and moreover, simulation studies are also conducted to show the usefulness of the GIVE statistic.

Credible School Value-Added with Undersubscribed School Lotteries

We introduce two empirical strategies harnessing the randomness in school assignment mechanisms to measure school value-added. The first estimator controls for the probability of school assignment, treating take-up as ignorable. We test this assumption using randomness in assignments. The second approach uses assignments as instrumental variables (IVs) for low-dimensional models of value-added and forms empirical Bayes posteriors from these IV estimates. Both strategies solve the underidentification challenge arising from school undersubscription. Models controlling for assignment risk and lagged achievement in Denver and New York City yield reliable value-added estimates. Estimates from models with lower-quality achievement controls are improved by IV.

Emigration and Political Contestation

How does migration affect global patterns of political violence and protest? While political scientists have examined the links between trade and conflict, less attention has been paid to the links between migration and conflict. In this paper, we show that greater emigration reduces domestic political violence by providing exit opportunities for aggrieved citizens and economic benefits to those who remain. Emigration also reduces non-violent forms of political contestation, including protests and strikes, implying that high emigration rates can produce relatively quiescent populations. However, larger flows of emigrants to democracies can increase non-violent protest in autocracies, as exposure to freer countries spreads democratic norms and the tools of peaceful opposition. We use instrumental variables analysis to account for the endogeneity of migration flows and find robust results for a range of indicators of civil violence and protest from 1960 to 2010.

Identification and application of plasmatic microRNA expression quantitative trait loci (miR-QTL) at first trimester of pregnancy

BackgroundMicroRNAs (miRNAs) are a class of small non-coding RNAs regulating gene expression. They are involved in many biological processes, including adaptation to pregnancy. The identification of genetic variants associated with gene expression, known as expression quantitative trait loci (eQTL), helps to understand the underlying molecular mechanisms and determinants of complex diseases. Using data from the prospective pre-birth Gen3G cohort, we investigated associations between maternal genotypes and plasmatic miRNA levels measured during the first trimester of pregnancy of 369 women.ResultsAssessing the associations between about 2 million SNPs and miRNA proximal pairs using best practices from the GTEx consortium, a total of 22,140 significant eQTLs involving 147 unique miRNAs were identified. Elastic-net regressions were applied to select the most relevant SNPs to build genetic risk scores (GRS) for each of these 147 miRNAs. For about half of the circulating miRNAs, the GRS captured >10% of the variance abundance. As a demonstration of the usefulness of the identified eQTLs and derived GRS, we used the GRSs as instrumental variables to test for association between the circulating levels of miRNAs quantified before the 16th week of pregnancy and the development of pregnancy complications (gestational diabetes [GDM] or pre-eclampsia [PE]) developing more than three months later on average. Using predicted miRNA levels derived from instrumental variables, we found 18 significant associations of miRNAs with potential support of causal inference for GDM or PE.ConclusionsOur results represent a valuable resource to understand miRNA regulation and highlight the potential of genetic instruments in predicting circulating miRNA levels and their possible contribution in disease development.