Instrumental variable estimation in ordinal probit models with mismeasured predictors

Abstract

Researchers in the medical, health, and social sciences routinely encounter ordinal variables such as self‐reports of health or happiness. When modelling ordinal outcome variables, it is common to have covariates, for example, attitudes, family income, retrospective variables, measured with error. As is well known, ignoring even random error in covariates can bias coefficients and hence prejudice the estimates of effects. We propose an instrumental variable approach to the estimation of a probit model with an ordinal response and mismeasured predictor variables. We obtain likelihood‐based and method of moments estimators that are consistent and asymptotically normally distributed under general conditions. These estimators are easy to compute, perform well and are robust against the normality assumption for the measurement errors in our simulation studies. The proposed method is applied to both simulated and real data. The Canadian Journal of Statistics 47: 653–667; 2019 © 2019 Statistical Society of Canada

Career Aspiration Fulfillment and COVID-19 Vaccination Intention among Nigerian Youth: An Instrumental Variable Approach

“…If Cov ( i ) = 0, endogeneity is absence. If an endogeneity problem [ Cov ( i ) ≠ 0] is present, the instrumental variable must be engaged in the specification of the vaccination model [ 48 , 49 , 50 , 51 ]. The dream job model is stated as: …”

Bayesian instrumental variable estimation in linear measurement error models

“…One of the commonly used methods of dealing with measurement error is the instrumental variable (IV) approach (e.g., Abarin et al, 2014; Abarin & Wang, 2012; Buzas & Stefanski, 1996; Carroll & Stefanski, 1994; Fuller, 1987; Guan et al, 2019; Guan & Wang, 2017; Wang, 2021; Wang & Hsiao, 2007, 2011; Xu, Ma & Wang, 2015). In practice, any variable that is correlated with the error‐prone covariates but is independent of the measurement error and model error can serve as a valid IV, e.g., a second independent measurement, or repeated measurements at different time points in longitudinal studies.…”