35333 research outputs found
Sort by
Local Overidentification And Efficiency Gains In Modern Causal Inference And Data Combination
No full textThis paper studies nonparametric local (over-)identification, in the sense of Chen and Santos (2018), and the associated semiparametric efficiency in modern causal frameworks. We develop a unified approach that begins by translating structural models with latent variables into their induced statistical models of observables and then analyzes local overidentification through conditional moment restrictions. We apply this approach to three leading models: (i) the general treatment model under unconfoundedness, (ii) the negative control model, and (iii) the long-term causal inference model under unobserved confounding. The first design yields a locally just-identified statistical model, implying that all regular asymptotically linear estimators of the treatment effect share the same asymptotic variance, equal to the (trivial) semiparametric efficiency bound. In contrast, the latter two models involve nonparametric endogeneity and are naturally locally overidentified; consequently, some doubly robust orthogonal moment estimators of the average treatment effect are inefficient. Whereas existing work typically imposes strong conditions to restore just-identification before deriving the efficiency bound, we relax such assumptions and characterize the general efficiency bound, along with efficient estimators, in the overidentified models (ii) and (iii)
Probability Pricing
No full textThis paper develops probability pricing, extending cash flow pricing to quantify the willingnessto-pay for changes in probabilities. We show that the value of any marginal change in probabilities can be expressed as a standard asset-pricing formula with hypothetical cash flows derived from changes in the survival function. This equivalence between probability and cash flow valuation allows us to construct hedging strategies and systematically decompose individual and aggregate willingness-to-pay. Four applications examine the valuation of changes in the distribution of aggregate consumption, the efficiency effects of varying performance precision in principal-agent problems, and the welfare implications of public and private information
Bach against Modernity
Full text linkA book review is presented for Bach against Modernity, authored by Michael Marissen, by Mark Peters
Forward Selection Fama-MacBeth Regression with Higher-Order Asset Pricing Factors
Full text linkWe show that the higher-orders and their interactions of the common sparse linear factors can effectively subsume the factor zoo. To this extend, we propose a forward selection Fama-MacBeth procedure as a method to estimate a high-dimensional stochastic discount factor model, isolating the most relevant higher-order factors. Applying this approach to terms derived from six widely used factors (the Fama-French five-factor model and the momentum factor), we show that the resulting higher-order model with only a small number of selected higher-order terms significantly outperforms traditional benchmarks both in-sample and out-of-sample. Moreover, it effectively subsumes a majority of the factors from the extensive factor zoo, suggesting that the pricing power of most zoo factors is attributable to their exposure to higher-order terms of common linear factors
Brazil: Banco BTG Pactual Emergency Liquidity Program, 2015
Full text linkOn November 25, 2015, André Esteves, then CEO of Banco BTG Pactual, a large Brazilian investment bank, was arrested by Brazilian authorities in connection with suspected involvement in a corruption scandal. Although the arrest did not involve BTG in any capacity and Esteves was later acquitted, the company’s stock quickly collapsed and depositors and other creditors rushed to reduce their exposures to the company. Depositors withdrew certificates of bank deposits, which BTG relied on to fund its daily operations. By November 27, BTG shares had fallen 26%. On December 2, the top seven shareholders of BTG took control of the company by swapping their preferred shares for Esteves’s common shares. On December 4, the Credit Guarantee Fund (Fundo Garantidor de Créditos, FGC), Brazil’s private, self-funded deposit insurance organization created by the government, extended a secured credit line to BTG for up to 6 billion Brazilian reais (BRL; USD 1.6 billion) to bolster BTG’s liquidity, with the aim of contributing to national financial stability. The FGC—and, by extension, its members’ fees—funded the credit line. The FGC sought and received confirmation from the Banco Central do Brasil (BCB), the central bank, that the failure of the bank would pose potential systemic risks. The FGC authorized the credit line for five years. It was 120% collateralized by a portion of BTG’s loan portfolio and personally guaranteed by the top seven shareholders of BTG. As of December 31, 2015, BTG had drawn a total of BRL 5 billion from the line. As of October 19, 2016, BTG had repaid all outstanding balances from the credit line. While the BCB itself did not provide the credit line, it played an “active role” in arranging the FGC’s credit line to BTG and also in the change of control by the shareholders
Iceland: Kaupthing Emergency Liquidity Program, 2008
Full text linkFollowing the privatization of Iceland’s state-owned banks between 1998 and 2003, the three largest banks in Iceland—Glitnir, Landsbanki, and Kaupthing—grew rapidly, with consolidated assets increasing from 100% of Iceland’s GDP in 2004 to nearly 900% by the end of 2007. Initially, this growth was funded by debt issuances in the European medium-term note market; however, as cracks in the international financial system appeared in 2006, the banks turned to offering high-interest savings accounts through their foreign subsidiaries. Beginning in October 2006, Kaupthing launched “Kaupthing Edge,” an online savings and deposit platform operating in markets outside Iceland. When the United States investment bank Lehman Brothers failed in September 2008, Iceland’s three major banks began to fail as a result of a run on their foreign currency deposits. On October 6, 2008, in an effort to prevent the failure of Kaupthing, the Central Bank of Iceland (CBI) provided it with a EUR 500 million loan against all of its shares in its subsidiary FIH Erhversbank A/S. Although the duration of the loan was initially set for four days, Kaupthing failed in three, on October 9. Immediately following Kaupthing’s failure, the Icelandic Financial Supervisory Authority appointed a resolution committee, which, in tandem with the CBI, began searching for a buyer for FIH. Upon the transfer of FIH to a consortium of new owners in January 2011, the CBI recovered EUR 255 million—approximately half of the initial loan amount
Youth Views On Health Content, Ads, And Messages On Social Media
Full text linkThis study aims to investigate youth perspectives on nutrition and healthism-related content, promotions, and messaging on social media. With the growing usage and impact of social media, especially in younger generations, it is important to recognize how social media may influence types of exposures and messaging of nutrition, weight management, and supplement-related content that youth are exposed to, which can, in turn, influence their perceptions on health and healthy lifestyles. These perspectives can then initiate engagement in helpful or harmful health behaviors. Through a qualitative analysis of focus group and interview responses of 29 Massachusetts teenagers, emerging themes were about: disordered weight control behaviors (DWCB); supplement messaging; messaging surrounding unhealthy foods, including mukbangs and binge eating; participants’ trustworthiness of influencers; and the overarching role of social media culture in spurring behavior change as a result of exposure to this content. Discussions about content surrounding unhealthy foods and behaviors were omnipresent in focus groups and interviews. Adolescents expressed variable feelings after viewing this type of content, ranging from disgust to expressing a strong desire to try particular foods. This resulted in behavior changes related to their daily eating habits. Many participants discussed that mistrust, misinformation, and unrealistic expectations from the food, beverage, or supplement influenced their opinions on these posts and the level of engagement they had with the product in their own lives. These findings can prompt future efforts to encourage brand and influencer transparency in product promotions, and social media platforms in reducing misinformation to their audiences
Characterization And Survival Outcome Of Missing Ecog Performance Status In A Real-World Cohort Of Advanced Non-Small Cell Lung Cancer Patients
Full text linkBackgroundThe Eastern Cooperative Oncology Group (ECOG) performance status (PS) is a critical factor in treatment decisions and prognosis for advanced non-small cell lung cancer (NSCLC). However, documentation gaps exist in clinical practice. The characteristics of patients with missing ECOG PS and the prognostic implications of this missingness are poorly understood. This study aimed to identify demographic and clinical predictors associated with missing ECOG PS documentation and to evaluate whether missing ECOG PS is associated with overall survival (OS) in a large real-world cohort of advanced NSCLC patients.
MethodsThis retrospective cohort study utilized the Flatiron Health electronic health record-derived, de-identified database. The cohort included 55,775 patients diagnosed with advanced NSCLC between 2011 and 2023 who initiated first-line therapy within 90 days of diagnosis. ECOG PS at treatment initiation was categorized as 0–1, 2–4, or missing. Multivariable logistic regression identified independent predictors of missing ECOG status. Kaplan-Meier methods and Cox proportional hazards regression (tested for and addressing non-proportionality) were used to compare OS among the three ECOG groups (Missing, 0-1, 2-4).
ResultsECOG PS was missing in 19.7% (n=10,993) of patients. The proportion of missing ECOG decreased significantly over time, from 50.2% in 2011 to 14.7% in 2023. Independent predictors of higher odds of missing ECOG included earlier diagnosis period (2011-2014 reference), Other race (OR 1.70 vs. White), older age (OR 1.01 per year), treatment in the West (OR 3.78 vs. Midwest) or South (OR 1.14 vs. Midwest) regions, academic practice setting (Community practice OR 0.53 vs. Academic), NSCLC NOS histology (OR 1.24 vs. Non-squamous), unknown/not documented (OR 9.81 vs. History) or no history of smoking (OR 1.09 vs. History), Medicaid (OR 1.13 vs. Commercial) or Other insurance (OR 1.15 vs. Commercial), and \u27No Testing\u27 for biomarkers (OR 2.04 vs. PD-L1 \u3c1%) or ALK Positive status (OR 1.24 vs. PD-L1 \u3c1%). Asian race (OR 0.73 vs. White), Medicare insurance (OR 0.82 vs. Commercial), and Squamous histology (OR 0.75 vs. Non-squamous) were associated with lower odds. Median OS was 18.4 months for ECOG 0-1, 12.0 months for ECOG 2-4, and 8.3 months for Missing ECOG. The multivariate logistic regression for 2-year survival showed that, compared to ECOG 2-4, ECOG 0-1 was associated with higher odds of survival (OR 1.49, 95% CI: 1.41-1.57), while Missing ECOG was associated with significantly lower odds of survival (OR 0.53, 95% CI: 0.49-0.57).
ConclusionWhile ECOG documentation in advanced NSCLC has improved, missingness persists (19.7% overall) and is associated with specific patient and system factors, including race, geographic region, practice setting, and lack of biomarker testing. Crucially, patients with missing ECOG exhibited the poorest survival outcomes, worse even than those documented as ECOG 2-4, both in unadjusted median survival and adjusted 2-year survival odds. Addressing documentation gaps through improved workflows and standardized data collection, particularly in identified high-risk groups and settings, is crucial for equitable care, reducing bias in real-world evidence, and appropriately accounting for the significant negative prognostic implication of missing performance status in oncology research
Nativity Status And Tobacco Use Among Hispanic Adults In The Us
Full text linkBackground: Tobacco use is the leading cause of preventable mortality. Understanding the consumption patterns of cigarettes and e-cigarettes, two of the most prevalent and harmful tobacco products, is important for population health. Furthermore, as the nation\u27s largest racial or ethnic minority group, the Hispanic population faces tobacco-related disparities despite overall low prevalence of tobacco use. Since the Hispanic population is not homogeneous, it is important to disaggregate it by ancestry and nativity status to better guide policy and targeted interventions.
Methods: The 2022 September Tobacco Use Supplement to the Current Population Survey (TUS-CPS) included 37,540 participants in the US. Rao-Scott chi-squared tests and multivariate logistic regressions were used to analyze the associations between cigarette and e-cigarette use and nativity status for individuals of Mexican, Puerto Rican, Central American, South American, and Other Spanish ancestry. In addition, multivariate logistic regressions assessed cigarette and e-cigarette use across race, ethnicity, and Hispanic ancestral origin within the US-born population.
Results: After accounting for sociodemographic differences, only individuals of Mexican ancestry demonstrated differences in tobacco use by nativity status. US-born individuals had higher prevalences of current smoking and vaping, as well as daily vaping, and were more likely to allow cigarette or e-cigarette use at home than their foreign-born peers (p \u3c 0.05 in all cases). However, US-born individuals of Mexican ancestry were still at lower risk compared to US-born NH-Whites.
Conclusions: The failure to observe differences in tobacco use by nativity status, or differences with US-born NH-Whites among other Hispanic subgroups, may be attributed to their high risk or the small sample sizes limiting the statistical power needed to detect true differences. Future efforts should focus on better collection of Hispanic-specific data