scholarly journals Evaluating the impact of misclassification when estimating heterogeneous causal effects

2020 ◽  
Author(s):  
Wen Wei Loh ◽  
Jee-Seon Kim
Keyword(s):  
Specific Effect ◽  
Parametric Bootstrap ◽  
Causal Effects ◽  
Sensitivity Analyses ◽  
Voter Mobilization ◽  
Heterogeneous Treatment Effects ◽  
Population Inference ◽  
The Individual ◽  
Effect Heterogeneity ◽  
The Impact

There is increasing attention given to assessing treatment effect heterogeneity arising from individuals belonging to different underlying classes in the population. Inference proceeds by separating the individuals into distinct classes, then estimating the causal effects within each class. In practice, the individual class memberships are rarely known with certainty, and often have to be estimated. Ignoring the uncertainty in the assumed class memberships precludes the possibility of misclassification, which can potentially lead to biased results and incorrect conclusions. In this paper, we propose a strategy for conducting sensitivity analyses to possible misclassification when estimating heterogeneous treatment effects for different classes. We exploit each individual's (typically nonzero) estimated probabilities of belonging to any given class to evaluate the impact of changing the assumed class memberships - one individual at a time - on the resultant class-specific effect estimates. Because the estimated probabilities are themselves subject to sampling variability, we propose Monte Carlo bounds that explicitly reflect the uncertainty in the individual class memberships via perturbations using a parametric bootstrap. We illustrate our proposed strategy using publicly available data from a field experiment with almost 11,000 voters to investigate whether the effect of voter mobilization on turnout varies across different voter classes. We demonstrate via simulation studies that the perturbed class membership probabilities may be used to construct confidence intervals that perform better empirically at attaining the nominal coverage rate, than existing methods that hold the estimated class memberships fixed.

scholarly journals Global sensitivity analyses of a neural networks model for a flotation circuit

2020 ◽  
Vol 74 (4) ◽  
pp. 247-256
Author(s):  
Manuel Saldaña ◽  
Luis Ayala ◽  
David Torres ◽  
Norman Toro
Keyword(s):  
Gas Flow ◽  
Particle Density ◽  
Machine Learning Algorithms ◽  
Sensitivity Analyses ◽  
Bubble Velocity ◽  
Main Process ◽  
Operational Parameters ◽  
Global Sensitivity ◽  
The Individual ◽  
The Impact

Modeling of flotation processes is complex due to the large number of variables involved and the lack of knowledge on the impact of operational parameters on the response(s), and given this problem, machine learning algorithms emerge as an alternative interesting when modeling dynamic processes. In this work, different artificial neural network (ANN) architectures for modeling the mineral concentrate in a rougher-cleaner-scavenger (RCS) circuit based on the main process variables are generated (variables as the recovery of the rougher, cleaner and scavenger cells, along with disaggregated variables). Analysis of the global sensitivity was performed to study the importance of the individual and joint performances of the stages of the flotation circuit, reflected by sensitivity indicators that allow to infer the impact that the stages and operational parameters produce on the dependent variables (mineral concentrate in rougher, cleaner and scavenger cells, in addition to the global concentration in the RCS circuit). It should be noted that the ANN is a useful tool for modeling dynamic systems such as flotation, while sensitivity analysis shows that the operation of the three threads turns out to be crucial for the subsequent evaluation of the circuit, while the Unbundled variables that most interact with the overall recovery are gas flow rate, bubble and particle diameters, bubble velocity, particle density, and surface tension.

Uncovering Sociological Effect Heterogeneity Using Tree-Based Machine Learning

2021 ◽  
pp. 008117502199350
Author(s):  
Jennie E. Brand ◽  
Jiahui Xu ◽  
Bernard Koch ◽  
Pablo Geraldo
Keyword(s):  
Machine Learning ◽  
Decision Trees ◽  
Nearest Neighbor ◽  
Recursive Partitioning ◽  
Specific Effect ◽  
Sensitivity Analyses ◽  
Similar Data ◽  
Race And Gender ◽  
Doubly Robust ◽  
Effect Heterogeneity

Individuals do not respond uniformly to treatments, such as events or interventions. Sociologists routinely partition samples into subgroups to explore how the effects of treatments vary by selected covariates, such as race and gender, on the basis of theoretical priors. Data-driven discoveries are also routine, yet the analyses by which sociologists typically go about them are often problematic and seldom move us beyond our biases to explore new meaningful subgroups. Emerging machine learning methods based on decision trees allow researchers to explore sources of variation that they may not have previously considered or envisaged. In this article, the authors use tree-based machine learning, that is, causal trees, to recursively partition the sample to uncover sources of effect heterogeneity. Assessing a central topic in social inequality, college effects on wages, the authors compare what is learned from covariate and propensity score–based partitioning approaches with recursive partitioning based on causal trees. Decision trees, although superseded by forests for estimation, can be used to uncover subpopulations responsive to treatments. Using observational data, the authors expand on the existing causal tree literature by applying leaf-specific effect estimation strategies to adjust for observed confounding, including inverse propensity weighting, nearest neighbor matching, and doubly robust causal forests. We also assess localized balance metrics and sensitivity analyses to address the possibility of differential imbalance and unobserved confounding. The authors encourage researchers to follow similar data exploration practices in their work on variation in sociological effects and offer a straightforward framework by which to do so.

scholarly journals Functional modeling of the craniospinal system for in-vitro parameter studies on the pathogenesis of NPH

2017 ◽  
Vol 3 (2) ◽  
pp. 825-828 ◽  
Author(s):  
Anne Benninghaus ◽  
Christine Goffin ◽  
Steffen Leonhardt ◽  
Klaus Radermacher
Keyword(s):  
Vascular System ◽  
Dynamic Compliance ◽  
Normal Pressure ◽  
The Elderly ◽  
Sensitivity Analyses ◽  
Common Disease ◽  
Novel Approach ◽  
The Individual ◽  
The Impact

AbstractNormal Pressure Hydrocephalus (NPH) has become a common disease in the elderly coming along with typical symptoms of dementia, gait ataxia and urinary incontinence, which make the differential diagnosis with other forms of dementia difficult. Furthermore the pathogenesis of NPH is still not understood. About 10% of all demented patients might be suffering from NPH [1]. Many hypotheses suggest that modified biomechanical boundary conditions affect the craniospinal dynamics inducing the pathogenesis of NPH. We present a novel approach for an in-vitro model of the craniospinal system to investigate important hydrodynamic influences on the system such as (dynamic) compliance of the vascular system and especially the spinal subarachnoid space (SAS) as well as reabsorption and hydrostatics. The experimental set-up enables the individual adjustment of relevant parameters for sensitivity analyses regarding the impact of resulting CSF dynamics on the pathogenesis of NPH.

An omnibus approach to assess covariate balance in observational studies using the distance covariance

2019 ◽  
Vol 29 (7) ◽  
pp. 1846-1866 ◽  
Author(s):  
Adin-Cristian Andrei ◽  
Patrick M McCarthy
Keyword(s):  
Observational Studies ◽  
Type I Error ◽  
Simulated Data ◽  
Causal Effects ◽  
Type I ◽  
Omnibus Test ◽  
Covariate Balance ◽  
Distance Covariance ◽  
The Individual ◽  
The Impact

Adequate baseline covariate balance among groups is critical in observational studies designed to estimate causal effects. Propensity score-based methods are popular ways to achieve covariate balance among groups. Existing methods are not easily generalizable to situations in which covariates of mixed type are collected nor do they provide a convenient way to compare the overall covariate vector distributions. Instead, covariate balance is assessed at the individual covariate level, thus the potential for increased overall type I error. We propose the use of the distance covariance, developed by Székely and colleagues, as an omnibus test of independence between covariate vectors and study group. We illustrate the advantages of this methodology in simulated data and in a cardiac surgery study designed to assess the impact of preoperative statin therapy on outcomes.

scholarly journals The Mean Unfulfilled Lifespan (MUL): A new indicator of the impact of mortality shocks on the individual lifespan, with application to mortality reversals induced by COVID-19

PLoS ONE ◽  
2021 ◽  
Vol 16 (7) ◽  
pp. e0254925
Author(s):  
Patrick Heuveline
Keyword(s):  
Mexico City ◽  
Sensitivity Analyses ◽  
Age At Death ◽  
Small Areas ◽  
Fine Grained ◽  
Life Expectancy At Birth ◽  
The Mean ◽  
The Difference ◽  
The Individual ◽  
The Impact

Declines in period life expectancy at birth (PLEB) provide seemingly intuitive indicators of the impact of a cause of death on the individual lifespan. Derived under the assumption that future mortality conditions will remain indefinitely those observed during a reference period, however, their intuitive interpretation becomes problematic when period conditions reflect a temporary mortality “shock”, resulting from a natural disaster or the diffusion of a new epidemic in the population for instance. Rather than to make assumptions about future mortality, I propose measuring the difference between a period average age at death and the average expected age at death of the same individuals (death cohort): the Mean Unfulfilled Lifespan (MUL). For fine-grained tracking of the mortality impact of an epidemic, I also provide an empirical shortcut to MUL estimation for small areas or short periods. For illustration, quarterly MUL values in 2020 are derived from estimates of COVID-19 deaths that might substantially underestimate overall mortality change in affected populations. These results nonetheless illustrate how MUL tracks the mortality impact of the pandemic in several national and sub-national populations. Using a seven-day rolling window, the empirical shortcut suggests MUL peaked at 6.43 years in Lombardy, 8.91 years in New Jersey, and 6.24 years in Mexico City for instance. Sensitivity analyses are presented, but in the case of COVID-19, the main uncertainty remains the potential gap between reported COVID-19 deaths and actual increases in the number of deaths induced by the pandemic in some of the most affected countries. Using actual number of deaths rather than reported COVID-19 deaths may increase seven-day MUL from 6.24 to 8.96 years in Mexico City and from 2.67 to 5.49 years in Lima for instance. In Guayas (Ecuador), MUL is estimated to have reached 12.7 years for the entire month of April 2020.

Accent Policy and Accent Modification Enterprises as Potential Indicators of Intercultural Power Relations: A Call for an Updated Research Agenda

2013 ◽  
Vol 3 (1) ◽  
pp. 22-33 ◽  
Author(s):  
Brynne D. Ovalle ◽  
Rahul Chakraborty
Keyword(s):  
International Relations ◽  
Identity Construction ◽  
Social Theory ◽  
Power Relations ◽  
Linguistic Diversity ◽  
Cultural Organization ◽  
Using Data ◽  
The Individual ◽  
The Impact ◽  
The Relationship

This article has two purposes: (a) to examine the relationship between intercultural power relations and the widespread practice of accent discrimination and (b) to underscore the ramifications of accent discrimination both for the individual and for global society as a whole. First, authors review social theory regarding language and group identity construction, and then go on to integrate more current studies linking accent bias to sociocultural variables. Authors discuss three examples of intercultural accent discrimination in order to illustrate how this link manifests itself in the broader context of international relations (i.e., how accent discrimination is generated in situations of unequal power) and, using a review of current research, assess the consequences of accent discrimination for the individual. Finally, the article highlights the impact that linguistic discrimination is having on linguistic diversity globally, partially using data from the United Nations Educational, Scientific and Cultural Organization (UNESCO) and partially by offering a potential context for interpreting the emergence of practices that seek to reduce or modify speaker accents.

Investigating the Relationship Between Self-Esteem and Stigma Among Young Adults With History of Suicide Attempts

Crisis ◽  
2016 ◽  
Vol 37 (4) ◽  
pp. 265-270 ◽  
Author(s):  
Meshan Lehmann ◽  
Matthew R. Hilimire ◽  
Lawrence H. Yang ◽  
Bruce G. Link ◽  
Jordan E. DeVylder
Keyword(s):  
Young Adults ◽  
Suicide Attempts ◽  
Self Esteem ◽  
High Risk Population ◽  
High Functioning ◽  
People With Mental Illness ◽  
History Of ◽  
The Individual ◽  
The Impact ◽  
The Relationship

Abstract. Background: Self-esteem is a major contributor to risk for repeated suicide attempts. Prior research has shown that awareness of stigma is associated with reduced self-esteem among people with mental illness. No prior studies have examined the association between self-esteem and stereotype awareness among individuals with past suicide attempts. Aims: To understand the relationship between stereotype awareness and self-esteem among young adults who have and have not attempted suicide. Method: Computerized surveys were administered to college students (N = 637). Linear regression analyses were used to test associations between self-esteem and stereotype awareness, attempt history, and their interaction. Results: There was a significant stereotype awareness by attempt interaction (β = –.74, p = .006) in the regression analysis. The interaction was explained by a stronger negative association between stereotype awareness and self-esteem among individuals with past suicide attempts (β = –.50, p = .013) compared with those without attempts (β = –.09, p = .037). Conclusion: Stigma is associated with lower self-esteem within this high-functioning sample of young adults with histories of suicide attempts. Alleviating the impact of stigma at the individual (clinical) or community (public health) levels may improve self-esteem among this high-risk population, which could potentially influence subsequent suicide risk.

The Role of Threat and Economic Uncertainty in Support for Scottish Independence

2014 ◽  
Vol 23 (1) ◽  
pp. 103-124 ◽  
Author(s):  
Daniel Kopasker
Keyword(s):  
Economic Consequences ◽  
Perceived Threat ◽  
Constitutional Change ◽  
Lack Of Information ◽  
Economic Expectations ◽  
Scottish Independence ◽  
The Individual ◽  
The Impact ◽  
Levels Of Support

Existing research has consistently shown that perceptions of the potential economic consequences of Scottish independence are vital to levels of support for constitutional change. This paper attempts to investigate the mechanism by which expectations of the economic consequences of independence are formed. A hypothesised causal micro-level mechanism is tested that relates constitutional preferences to the existing skill investments of the individual. Evidence is presented that larger skill investments are associated with a greater likelihood of perceiving economic threats from independence. Additionally, greater perceived threat results in lower support for independence. The impact of uncertainty on both positive and negative economic expectations is also examined. While uncertainty has little effect on negative expectations, it significantly reduces the likelihood of those with positive expectations supporting independence. Overall, it appears that a general economy-wide threat is most significant, and it is conjectured that this stems a lack of information on macroeconomic governance credentials.

Laughter on the Fringes

2019 ◽  
Author(s):  
Anna Peterson
Keyword(s):  
The Other ◽  
Literary Canon ◽  
Autobiographical Writing ◽  
Treasure Trove ◽  
Old Comedy ◽  
Other Hand ◽  
Aggressive Humor ◽  
The One ◽  
The Individual ◽  
The Impact

This book examines the impact that Athenian Old Comedy had on Greek writers of the Imperial era. It is generally acknowledged that Imperial-era Greeks responded to Athenian Old Comedy in one of two ways: either as a treasure trove of Atticisms, or as a genre defined by and repudiated for its aggressive humor. Worthy of further consideration, however, is how both approaches, and particularly the latter one that relegated Old Comedy to the fringes of the literary canon, led authors to engage with the ironic and self-reflexive humor of Aristophanes, Eupolis, and Cratinus. Authors ranging from serious moralizers (Plutarch and Aelius Aristides) to comic writers in their own right (Lucian, Alciphron), to other figures not often associated with Old Comedy (Libanius) adopted aspects of the genre to negotiate power struggles, facilitate literary and sophistic rivalries, and provide a model for autobiographical writing. To varying degrees, these writers wove recognizable features of the genre (e.g., the parabasis, its agonistic language, the stage biographies of the individual poets) into their writings. The image of Old Comedy that emerges from this time is that of a genre in transition. It was, on the one hand, with the exception of Aristophanes’s extant plays, on the verge of being almost completely lost; on the other hand, its reputation and several of its most characteristic elements were being renegotiated and reinvented.

Sign in / Sign up

Export Citation Format

Share Document