ABSTRACT
This chapter shows that it is possible to estimate the intercept of a structural equation with ordinary multiple regression, and to use this information to fully specify and estimate the structural model. Since the pioneer work of D. A. Kenny and C. M. Judd, two problems have precluded the wide use of structural equation models that include nonlinear relations among latent continuous variables. The first problem is that most, if not all, of the methods that have been proposed to specify this type of nonlinear relations use products of observed variables as multiple indicators of the product term of the latent independent variables involved in the nonlinear relation. The second problem is more fundamental and stems from the nature of nonlinear relations among continuous variables. The chapter deals with these two problems, and we propose a solution for each of them.
