Binary outcomes and endogenous explanatory variables: tests and solutions with an application to the demand for contraceptive use in Tunisia.
A simple analytic framework for estimating the effects of explanatory variables on discrete outcomes is presented that controls for the potential endogeneity of explanatory variables. In addition, tests for exogeneity and joint determination of the outcomes and the explanatory variables are outline...
| Publicado en: | Demography Vol. 32; pp. 111 - 132 |
|---|---|
| Autores principales: | , , |
| Formato: | Artículo |
| Publicado: |
Springer Science & Business Media B.V.
February 1995
|
| Materias: | |
| Acceso en línea: | Ver este registro en EBSCOhost |
| Sumario: | A simple analytic framework for estimating the effects of explanatory variables on discrete outcomes is presented that controls for the potential endogeneity of explanatory variables. In addition, tests for exogeneity and joint determination of the outcomes and the explanatory variables are outlined. The findings of a Monte Carlo study of the performance of these techniques are summarized, and these findings are employed to suggest how researchers should approach these problems in practice. Data from the 1988 Tunisia Demographic and Health Survey are analyzed using these methods in order to ascertain the effect of fertility intentions on contraceptive use. The unexpected result of the Monte Carlo study that the simple probit and two-step probit regression methods often perform as well as much more complicated methods is noted. Some specific recommendations are provided. |
|---|