Nonrecursive Causal Models, Issue 37Where an assumption of unidirectionality in causal effects is unrealistic, 'recursive' models cannot be used, and more complex 'nonrecursive' models are necessary. Unfortunately, many nonrecursive models (unlike recursive models) are 'unidentified', which makes meaningful parameter estimation impossible. Even when they are identified, it would be inappropriate to use OLS regression techniques (appropriate for recursive models) for the purpose of estimation. The concept of identification, and the factors that lead to it are explained; and various tests for determination are provided. Illustrations from a variety of social science disciplines are used throughout the book. |
Contents
Series Editors Introduction | 5 |
The Identification Problem | 18 |
Testing for Identification | 39 |
Copyright | |
4 other sections not shown
Other editions - View all
Common terms and phrases
2SLS estimates achieve identification algorithm assumed block recursive models Causal Diagram causal effects chosen infinite sample coefficients collapsed matrix Comparative Candidate Evaluation correlation deleting demand curve determine direct effect equation is identified equations 22 equations 61 Equations 80 error terms estimating the parameters exactly identified equations exogenous explanatory variables full-information high multicollinearity indirect least squares instrumental variables Kritzer lagged endogenous variable level-curve linear combination m₁ matrix algebra model is identified Model of Equations model of Figure multiequation model nonidentified models nonrecursive model OLS estimates order condition parameters of equations party ID pivot column pivot row policy/issue positions procedure properly chosen infinite protest violence model Protestors R² values reduced-form equations reduced-form parameters reduced-form perspective regression analysis second-stage regression set of parameters simple form social science standard errors structural equation structural parameters supply curve system matrix terms are uncorrelated test for identification test row tion voting behavior model X₁ X₂