Residual diagnostics for mixture models

摘要: Abstract A sample is commonly modeled by a mixture distribution if the observations follow common distribution, but parameter of interest differs between observations. For example, we observe lengths not ages offish. It may be reasonable to assume that length normally distributed about an unknown mean depends on age fish. Provided there more than one class in sample, then data are as normals. In this article random from exponential family distributions and for each observation sampled independently mixing Q. The adequacy fitted model can assessed examining residuals based ratio observed expected fit. Residuals homogeneity (in which Q one-point distribution) display convexity property when model; becomes basis fo...