scholarly journals Sensitivity Analysis for Causal Inference under Unmeasured Confounding and Measurement Error Problems

2013 ◽  
Vol 9 (2) ◽  
Author(s):  
Iván Díaz ◽  
Mark J. van der Laan
Keyword(s):  
Sensitivity Analysis ◽  
Measurement Error ◽  
Causal Inference ◽  
Parametric Model ◽  
Linear Estimator ◽  
Nonparametric Model ◽  
Asymptotically Linear ◽  
Full Data ◽  
Unmeasured Confounding ◽  
New Model

AbstractIn this article, we present a sensitivity analysis for drawing inferences about parameters that are not estimable from observed data without additional assumptions. We present the methodology using two different examples: a causal parameter that is not identifiable due to violations of the randomization assumption, and a parameter that is not estimable in the nonparametric model due to measurement error. Existing methods for tackling these problems assume a parametric model for the type of violation to the identifiability assumption and require the development of new estimators and inference for every new model. The method we present can be used in conjunction with any existing asymptotically linear estimator of an observed data parameter that approximates the unidentifiable full data parameter and does not require the study of additional models.

scholarly journals Methodological Challenges When Studying Distance to Care as an Exposure in Health Research

2019 ◽  
Vol 188 (9) ◽  
pp. 1674-1681 ◽  
Author(s):  
Ellen C Caniglia ◽  
Rebecca Zash ◽  
Sonja A Swanson ◽  
Kathleen E Wirth ◽  
Modiegi Diseko ◽  
...  
Keyword(s):  
Measurement Error ◽  
Causal Inference ◽  
Selection Bias ◽  
Birth Outcomes ◽  
Health Research ◽  
Instrumental Variable ◽  
Sensitivity Analyses ◽  
Unmeasured Confounding ◽  
Instrument Selection ◽  
Distance To Care

Abstract Distance to care is a common exposure and proposed instrumental variable in health research, but it is vulnerable to violations of fundamental identifiability conditions for causal inference. We used data collected from the Botswana Birth Outcomes Surveillance study between 2014 and 2016 to outline 4 challenges and potential biases when using distance to care as an exposure and as a proposed instrument: selection bias, unmeasured confounding, lack of sufficiently well-defined interventions, and measurement error. We describe how these issues can arise, and we propose sensitivity analyses for estimating the degree of bias.

Design for Reliability and Maintainability Cost Model

2013 ◽  
Author(s):  
Basel Alsayyed Ahmad
Keyword(s):  
Sensitivity Analysis ◽  
Product Development ◽  
Cost Model ◽  
Parametric Model ◽  
Design Stage ◽  
Cost Models ◽  
Product Lifecycle ◽  
New Model ◽  
Design For Reliability ◽  
Design For Maintainability

In this paper, the author presents a cost model which simultaneously incorporates the abilities of two main designs. Namely, design for reliability (DfR) and design for maintainability (DfMn). DfR and DfMn have been of a great concern for designers. The focus has been more on the quality rather than aspects of reliability and maintainability of product development in general and product design in particular. However, as more awareness and attention is paid to the whole product lifecycle, more product lifecycle X-ability needs to be considered upfront at the design stage. To develop a new model, cost models of design for reliability as well as design for maintainability have been considered and evaluated. This new model, which tackles the two criteria simultaneously, has been developed and is a parametric model which allows for running sensitivity analysis based on the different aspects of reliability and maintainability.

scholarly journals Mendelian randomisation for mediation analysis: current methods and challenges for implementation

2021 ◽  
Author(s):  
Alice R. Carter ◽  
Eleanor Sanderson ◽  
Gemma Hammerton ◽  
Rebecca C. Richmond ◽  
George Davey Smith ◽  
...  
Keyword(s):  
Measurement Error ◽  
Causal Inference ◽  
Mediation Analysis ◽  
Instrumental Variable ◽  
Real Data ◽  
Mendelian Randomisation ◽  
Differential Measurement ◽  
Individual Level ◽  
Differential Measurement Error ◽  
Summary Data

AbstractMediation analysis seeks to explain the pathway(s) through which an exposure affects an outcome. Traditional, non-instrumental variable methods for mediation analysis experience a number of methodological difficulties, including bias due to confounding between an exposure, mediator and outcome and measurement error. Mendelian randomisation (MR) can be used to improve causal inference for mediation analysis. We describe two approaches that can be used for estimating mediation analysis with MR: multivariable MR (MVMR) and two-step MR. We outline the approaches and provide code to demonstrate how they can be used in mediation analysis. We review issues that can affect analyses, including confounding, measurement error, weak instrument bias, interactions between exposures and mediators and analysis of multiple mediators. Description of the methods is supplemented by simulated and real data examples. Although MR relies on large sample sizes and strong assumptions, such as having strong instruments and no horizontally pleiotropic pathways, our simulations demonstrate that these methods are unaffected by confounders of the exposure or mediator and the outcome and non-differential measurement error of the exposure or mediator. Both MVMR and two-step MR can be implemented in both individual-level MR and summary data MR. MR mediation methods require different assumptions to be made, compared with non-instrumental variable mediation methods. Where these assumptions are more plausible, MR can be used to improve causal inference in mediation analysis.

Is it possible to predict habitat use by spawning salmonids? A test using California golden trout (Oncorhynchus mykiss aguabonita)

1999 ◽  
Vol 56 (9) ◽  
pp. 1576-1584 ◽  
Author(s):  
Roland A Knapp ◽  
Haiganoush K Preisler
Keyword(s):  
Oncorhynchus Mykiss ◽  
Water Depth ◽  
Parametric Model ◽  
Water Velocity ◽  
Habitat Characteristics ◽  
Nonparametric Model ◽  
Habitat Features ◽  
Substrate Size ◽  
Depth Water ◽  
Spawning Sites

It is widely believed that stream salmonids select spawning sites based on water depth, water velocity, and substrate size. Attempts to predict spawning locations using these habitat features have met with little success, however. In this study, we used nonparametric logistic regression to determine what habitat features were associated with the locations chosen by spawning California golden trout (Oncorhynchus mykiss aguabonita). From this nonparametric model, we developed a parametric model that incorporated the habitat features most strongly associated with spawning sites and used this model to calculate the probability of use by spawning golden trout for specific stream locations. The overall nonparametric model was highly significant and explained 62% of the variation in spawning location. Four of the eight habitat variables, substrate size, water depth, water velocity, and stream width, had highly significant effects and alone explained 59% of the variation in spawning location. The results of a cross-validation procedure indicated that the parametric model generally provided a good fit to the data. These results indicate that location-specific probabilities of use were predictable based on easily measured habitat characteristics and that nonparametric regression, an approach still rarely used in ecological studies, may have considerable utility in the development of fish-habitat models. Given the escalating pace at which fish habitats are being altered, such models are increasingly important in predicting the effects of these alterations on populations.

Sign in / Sign up

Export Citation Format

Share Document