Estimation of transition rates in dynamic household models.

摘要: This chapter deals with the estimation of instantaneous transition rates in nonparametric semiparametric and parametric models. The bivariate multivariate problems are introduced discussed using data from a retrospective life history survey undertaken France 1981. exploration pinpoints some important issues for future research. An problem is that it was not possible to estimate complete models because this would require huge sample many observations each cell. Such detailed information available. Therefore authors assumptions on main effects mainly based earlier experiences demographic data. For example applies prominent effect duration since last event rather than ones. Another will be introduce more time-dependent explanatory variables into these analyses. When revisiting such variable namely being married or at time t. an approach needs generalized. A 3rd did appear clearly previous analyses but seems very previously different control their effect. However other characteristics studied individuals interfere them while there no possibility measuring characteristics. can incorporate unobserved heterogeneity research has only just begun. Lastly have "fuzzy time" leads which difficult deal with. precise scale events placed. when studying observe "fuzzy" time. It does seem working interval 6 months even year know whether birth migration occurred 1st.