A generalized recursive max-linear model on a directed graph is a structural equation model where each variable is a max-linear function of their parental node variables and independent noise variables. It is a multidimensional fixed point equation in a maxtimes algebra and its fixed point can be approached by iteration. When time is introduced into this model, it becomes a deterministic max-times system. The stability of this system is of high interest and is investigated when the maximum cycle mean of the coefficient matrix is less than or equal to 1. Inspired by the fact that in a Gaussian graphical model conditional independence given all remaining random variables corresponds to a zero entry in the inverse covariance matrix, the relationship between conditional dependence structure and tail dependence structure for a recursive max-linear model on a polytree is also investigated.     «

A generalized recursive max-linear model on a directed graph is a structural equation model where each variable is a max-linear function of their parental node variables and independent noise variables. It is a multidimensional fixed point equation in a maxtimes algebra and its fixed point can be approached by iteration. When time is introduced into this model, it becomes a deterministic max-times system. The stability of this system is of high interest and is investigated when the maximum cycle...     »