scholarly journals A Graphical Approximation to Generalization: Definitions and Diagrams

2015 ◽  
Vol 6 (1) ◽  
Author(s):  
Fernando Martel García ◽  
Leonard Wantchekon
Keyword(s):  
External Validity ◽  
Fundamental Problem ◽  
Systematic Variation ◽  
Causal Effects ◽  
Graphical Language ◽  
Causal Diagrams ◽  
Causal Approach

AbstractThe fundamental problem of external validity is not to generalize from one experiment, so much as to experimentally test generalizable theories. That is, theories that explain the systematic variation of causal effects across contexts. Here we show how the graphical language of causal diagrams can be used in this endeavour. Specifically we show how generalization is a causal problem, how a causal approach is more robust than a purely predictive one, and how causal diagrams can be adapted to convey partial parametric information about interactions.

scholarly journals 1274The application of simulation to quantifying the influence of bias in reproductive and perinatal epidemiology

2021 ◽  
Vol 50 (Supplement_1) ◽  
Author(s):  
Jennifer Dunne ◽  
Gizachew Assefa Tessema ◽  
Milica Ognjenovic ◽  
Gavin Pereira
Keyword(s):  
Observational Studies ◽  
Reference Lists ◽  
Postnatal Period ◽  
Causal Effects ◽  
Simulation Methods ◽  
Perinatal Epidemiology ◽  
Causal Diagrams ◽  
Practical Guidance ◽  
The Many ◽  
Development Analysis

Abstract Background Establishing causal effects in reproductive and perinatal epidemiology is challenging due to the many selection and attrition processes from preconception to the postnatal period. Further challenging, is the potential for the misclassification of exposures, outcomes and confounders, contributing to measurement error. The application of simulation enables the illustration and quantification of the magnitude of various types of bias commonly found in observational studies. Methods A systematic search was conducted in PubMed, Medline, Embase, CINAHL and Scopus in August 2020. A gray literature search of Google and Google Scholar, followed by a search of the reference lists of included studies, was undertaken. Results Thirty-nine studies, covering information (n = 14), selection (n = 14), confounding (n = 9), protection (n = 1), and attenuation bias (n = 1) were identified. The methods of simulating data and reporting of results varied, with more recent studies including causal diagrams. Few studies included code for replication. Although there has been an increasing application of simulation in reproductive and perinatal epidemiology since 2015, overall this remains an underexplored area. Conclusions The studies demonstrated effectiveness in the quantification of multiple types of bias using simulation. The limited use implies that further effort is required to increase knowledge of the application of simulation, which will thereby improve causal interpretation in reproductive and perinatal studies. Key messages Practical guidance for researchers is required in the development, analysis and reporting of simulation methods for the quantification of bias.

Replicating Experiments Using Aggregate and Survey Data: The Case of Negative Advertising and Turnout

1999 ◽  
Vol 93 (4) ◽  
pp. 901-909 ◽  
Author(s):  
Stephen D. Ansolabehere ◽  
Shanto Iyengar ◽  
Adam Simon
Keyword(s):  
Survey Data ◽  
Instrumental Variables ◽  
External Validity ◽  
External Validation ◽  
Internal Validity ◽  
Aggregate Data ◽  
Causal Effects ◽  
Negative Advertising ◽  
Senate Elections ◽  
Advertising Exposure

Experiments show significant demobilizing and alienating effects of negative advertising. Although internally valid, experiments may have limited external validity. Aggregate and survey data offer two ways of providing external validation for experiments. We show that survey recall measures of advertising exposure suffer from problems of internal validity due to simultaneity and measurement error, which bias estimated effects of ad exposure. We provide valid estimates of the causal effects of ad exposure for the NES surveys using instrumental variables and find that negative advertising causes lower turnout in the NES data. We also provide a careful statistical analysis of aggregate turnout data from the 1992 Senate elections that Wattenberg and Brians (1999) recommend. These aggregate data confirm our original findings. Experiments, surveys, and aggregate data all point to the same conclusion: Negative advertising demobilizes voters.

scholarly journals Who’s in and Who’s Out? Selection Bias in Aging Research

2020 ◽  
Vol 4 (Supplement_1) ◽  
pp. 822-822
Author(s):  
Elizabeth Rose Mayeda ◽  
Eleanor Hayes-Larson ◽  
Hailey Banack
Keyword(s):  
Selection Bias ◽  
External Validity ◽  
Sample Selection ◽  
Internal Validity ◽  
Causal Effects ◽  
Aging Research ◽  
Selection Processes ◽  
The People ◽  
Biased Estimates ◽  
Internal And External Validity

Abstract Selection bias presents a major threat to both internal and external validity in aging research. “Selection bias” refers to sample selection processes that lead to statistical associations in the study sample that are biased estimates of causal effects in the population of interest. These processes can lead to: (1) results that do not generalize to the population of interest (threat to external validity) or (2) biased effect estimates (associations that do not represent causal effects for any population, including the people in the sample; a threat to internal validity). In this presentation, we give an overview of selection bias in aging research. We will describe processes that can give rise to selection bias, highlight why they are particularly pervasive in this field, and present several examples of selection bias in aging research. We end with a brief summary of strategies to prevent and correct for selection bias in aging research.

scholarly journals Counterfactual causal analysis on structured data

2021 ◽  
Author(s):  
swarna paul ◽  
Tauseef Jamal Firdausi ◽  
Saikat Jana ◽  
Arunava Das ◽  
Piyush Nandi
Keyword(s):  
Fundamental Problem ◽  
Business Environment ◽  
Structured Data ◽  
Causal Effects ◽  
High Fidelity ◽  
Causal Relations ◽  
Daunting Task ◽  
Data Graph ◽  
Business Events ◽  
Causal Variables

Data generated in a real-world business environment can be highly connected with intricate relationships among entities. Studying relationships and understanding their dynamics can provide deeper understanding of business events. However, finding important causal relations among entities is a daunting task with heavy dependency on data scientists. Also due to fundamental problem of causal inference it is impossible to directly observe causal effects. Thus, a method is proposed to explain predictive causal relations in an arbitrary linked dataset using counterfactual type causality. The proposed method can generate counterfactual examples with high fidelity in minimal time. It can explain causal relations among any chosen response variable and an arbitrary set of independent causal variables to provide explanations in natural language. The evidence of the explanations is shown in the form of a summarized connected data graph

scholarly journals The Necessity of Construct and External Validity for Generalized Causal Claims

2021 ◽  
Author(s):  
Kevin Esterling ◽  
David Brady ◽  
Eric Schwitzgebel
Keyword(s):  
Social Sciences ◽  
Construct Validity ◽  
External Validity ◽  
Causal Effect ◽  
Internal Validity ◽  
Causal Effects ◽  
Formal Framework ◽  
Tremendous Progress ◽  
The Social

The credibility revolution has facilitated tremendous progress in the social sciences by advancing design-based strategies that rely on internal validity to deductively identify causal effects. We demonstrate that prioritizing internal validity while neglecting construct and external validity prevents causal generalization and misleadingly converts a deductive claim of causality into a claim based on speculation and exploration -- undermining the very goals of the credibility revolution. We develop a formal framework of causal specification to demonstrate that internal, external and construct validity are jointly necessary for generalized claims regarding a causal effect. If one lacks construct validity, one cannot assign meaningful labels to the cause or to the outcome. If one lacks external validity, one cannot make statements about the conditions required for the cause to occur. Re-balancing considerations of internal, construct and external via causal specification preserves and advances the intent of the credibility revolution to understand causal effects.

GENERALIZED STATISTICAL COMPLEXITY MEASURE

2010 ◽  
Vol 20 (03) ◽  
pp. 775-785 ◽  
Author(s):  
OSVALDO A. ROSSO ◽  
LUCIANA DE MICCO ◽  
HILDA A. LARRONDO ◽  
MARÍA T. MARTÍN ◽  
ANGEL PLASTINO
Keyword(s):  
Dynamical System ◽  
Time Series ◽  
Probability Distribution ◽  
Metric Space ◽  
Fundamental Problem ◽  
Complexity Measure ◽  
Causal Effects ◽  
Probability Metric ◽  
Statistical Complexity ◽  
Selection Of

A generalized Statistical Complexity Measure (SCM) is a functional that characterizes the probability distribution P associated to the time series generated by a given dynamical system. It quantifies not only randomness but also the presence of correlational structures. We review here several fundamental issues in such a respect, namely, (a) the selection of the information measure [Formula: see text]; (b) the choice of the probability metric space and associated distance [Formula: see text]; (c) the question of defining the so-called generalized disequilibrium [Formula: see text]; (d) the adequate way of picking up the probability distribution P associated to a dynamical system or time series under study, which is indeed a fundamental problem. In this communication we show (point d) that sensible improvements in the final results can be expected if the underlying probability distribution is "extracted" via appropriate consideration regarding causal effects in the system's dynamics.

scholarly journals Counterfactual causal analysis on structured data

2021 ◽  
Author(s):  
swarna paul ◽  
Tauseef Jamal Firdausi ◽  
Saikat Jana ◽  
Arunava Das ◽  
Piyush Nandi
Keyword(s):  
Fundamental Problem ◽  
Business Environment ◽  
Structured Data ◽  
Causal Effects ◽  
High Fidelity ◽  
Causal Relations ◽  
Daunting Task ◽  
Data Graph ◽  
Business Events ◽  
Causal Variables

Data generated in a real-world business environment can be highly connected with intricate relationships among entities. Studying relationships and understanding their dynamics can provide deeper understanding of business events. However, finding important causal relations among entities is a daunting task with heavy dependency on data scientists. Also due to fundamental problem of causal inference it is impossible to directly observe causal effects. Thus, a method is proposed to explain predictive causal relations in an arbitrary linked dataset using counterfactual type causality. The proposed method can generate counterfactual examples with high fidelity in minimal time. It can explain causal relations among any chosen response variable and an arbitrary set of independent causal variables to provide explanations in natural language. The evidence of the explanations is shown in the form of a summarized connected data graph

scholarly journals Multi-Cause Effect Estimation with Disentangled Confounder Representation

2021 ◽  
Author(s):  
Jing Ma ◽  
Ruocheng Guo ◽  
Aidong Zhang ◽  
Jundong Li
Keyword(s):  
Real World ◽  
Fundamental Problem ◽  
Causal Effect ◽  
Causal Effects ◽  
Treatment Effect Estimation ◽  
Latent Representations ◽  
Real World Datasets ◽  
New Perspective ◽  
Multiple Treatments ◽  
Effect Estimation

One fundamental problem in causality learning is to estimate the causal effects of one or multiple treatments (e.g., medicines in the prescription) on an important outcome (e.g., cure of a disease). One major challenge of causal effect estimation is the existence of unobserved confounders -- the unobserved variables that affect both the treatments and the outcome. Recent studies have shown that by modeling how instances are assigned with different treatments together, the patterns of unobserved confounders can be captured through their learned latent representations. However, the interpretability of the representations in these works is limited. In this paper, we focus on the multi-cause effect estimation problem from a new perspective by learning disentangled representations of confounders. The disentangled representations not only facilitate the treatment effect estimation but also strengthen the understanding of causality learning process. Experimental results on both synthetic and real-world datasets show the superiority of our proposed framework from different aspects.

Sign in / Sign up

Export Citation Format

Share Document