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.

Reporting methodological issues of the mendelian randomization studies in health and medical research: a systematic review

Background Mendelian randomization (MR) studies using Genetic risk scores (GRS) as an instrumental variable (IV) have increasingly been used to control for unmeasured confounding in observational healthcare databases. However, proper reporting of methodological issues is sparse in these studies. We aimed to review published papers related to MR studies and identify reporting problems. Methods We conducted a systematic review using the clinical articles published between 2009 and 2019. We searched PubMed, Scopus, and Embase databases. We retrieved information from every MR study, including the tests performed to evaluate assumptions and the modelling approach used for estimation. Using our inclusion/exclusion criteria, finally, we identified 97 studies to conduct the review according to the PRISMA statement. Results Only 66 (68%) of the studies empirically verified the first assumption (Relevance assumption), and 40 (41.2%) studies reported the appropriate tests (e.g., R2, F-test) to investigate the association. A total of 35.1% clearly stated and discussed theoretical justifications for the second and third assumptions. 30.9% of the studies used a two-stage least square, and 11.3% used the Wald estimator method for estimating IV. Also, 44.3% of the studies conducted a sensitivity analysis to illuminate the robustness of estimates for violations of the untestable assumptions. Conclusions We found that incompleteness of the justification of the assumptions for the instrumental variable in MR studies was a common problem in our selected studies. This may misdirect the findings of the studies.

Structural Convergence between African Countries: Empirical Evidence

This paper aims to investigate structural convergence in selected African countries over the period 1994-2019. Using panel data for 48 African countries and several estimation methods [Panel-Corrected Standard Errors (PCSE), Feasible Generalized Least Squares (FGLS), tobit model, instrumental variable, and Granger non-causality], the results show the existence of the phenomenon of sectoral structural convergence in Africa, i.e. a greater similarity in sectoral structures while income gaps are narrowing. The paper also highlights the service sector's low relative productivity level and industrial sector's low labor force attractiveness despite a significant shift in labor from the agricultural sector and a higher level of relative productivity respectively. To address this issue, the development and acquisition of human and physical capital would be necessary to develop the industrial sector and increase the service sector's productivity.

Caring but sharing unintentionally: Lobbying for innovations and the leakage of knowledge

When firms engage in lobbying, their intended outcome is a regulatory change that benefits them. However, prior literature suggests that there may also be an unintended outcome of lobbying—the leakage of knowledge to competitors. In this paper, I explore when the intended and the unintended outcomes are more likely by theorizing about the relationship between lobbying and innovation. I predict that innovations that are novel are more likely to benefit from the intended regulatory changes. However, innovations that use knowledge uniquely possessed by a few firms are more likely to be compromised by the leakage of knowledge that happens during lobbying. I use new data from 1999-2013 on public U.S. firms that engaged in lobbying to federal agencies, the regulatory changes made by federal agencies, and the 16,000 patents applied for by those firms. I employ unsupervised machine learning (Doc2Vec) to measure knowledge leakage and an instrumental variable 2SLS mediation analyses to test the theory. The results suggest that the intended regulatory changes that follow lobbying can benefit innovations by facilitating wider adoption. However, unique technological knowledge that only a few firms possess may be expropriated by competitors during the process of lobbying. Overall, this paper demonstrates that fundamental aspects of innovation— such as institutional change, knowledge transfer, and technology adoption—are closely related to lobbying, a form of nonmarket activity.

Does Social Insurance Block the Inter-Generational Transmission of Poverty? Evidence from the 2018 China Family Panel Survey

In this research, we explored whether participation in pension insurance and medical insurance for children and fathers blocks the inter-generational transmission of poverty. Using data from the China Family Panel Survey of 2018, this paper took the average level of insurance participation of a sample group as an instrumental variable, applied the IV-probit model, and found that the participation of children in pension insurance and the participation of fathers in medical insurance significantly reduce the probability of the inter-generational transmission of poverty, but that the participation of children in medical insurance and the participation of fathers in pension insurance increase it. These results were robust. Furthermore, there was heterogeneity in household registration, geographical location, and marriage with regard to the impact of social insurance participation on the inter-generational transmission of poverty. These results could help the formulation of anti-poverty policies to address the inter-generational transmission of poverty.