Catchpole, E. A.; Morgan, B. J. T.; Viallefont, A. - In: Journal of Applied Statistics 29 (2002) 1-4, pp. 625-636
A model, involving a particular set of parameters, is said to be parameter redundant when the likelihood can be expressed in terms of a smaller set of parameters. In many important cases, the parameter redundancy of a model can be checked by evaluating the symbolic rank of a derivative matrix....