Interdependencies between events over an individual's life course are widespread. For example, the amount of schooling a person obtains may depend on his or her family behavior, whether he or she is married or not and the presence or absence of children. But vice versa, a person's family behavior may depend on his or educational behavior, whether he or she is in school or not. One of the most difficult problems facing research on interdependence between two or more life -spheres is the choice of statistical models for characterizing the dependencies. This problem is currently an obstacle to progress in the study of life-cycle behavior in tow or more life spheres. It is widely recognized that such interdependencies best should be studied by means of dynamic models, where the sequence and types of transitions are modelled as they actually occur over the life cycle: Through modelling of so-called event histories. A new such methodology was recently developed in Petersen. He developed an approach for analyzing two processes each of which has two states. It removed the weaknesses of the two existing event history approaches. The present project has five central goals. The first goal is to extend the approach in Petersen to two processes where each has three mor more states. The Second goal is to extend the approach in Petersen to more than two processes, each of which may have three or more states. The has clear relevance. In a number of cases there will be more than two processes moving simultaneously in time and where each of these may have more that two states. For example, a person may be subject to a marriage, child-bearing, and employment process. The third goal is to extend the approach in Petersen to the case where one or more of the processes has a continuous state space, which occurs in individual-level earnings and socioeconomic status histories. The fourth goal of the project is to address some of the conclusions from the theoretical work in Goals 1-3 in a series computer simulations. The fifth goal of the project is to assess how the new approach works with real-life data, using data from the National Longitudinal Survey of Youth, also comparing it to the tow existing approaches. Reaching these goals is relevant because the development of procedures for studying interdependencies among behaviors in various life spheres has been a long-standing unresolved issue among life-course researchers concerned for example with the interrelationships between child-bearing, marital behavior, family demographics, migration, and more.
