
handle: 11583/1404600
Summary: Computations with cumulants are becoming easier through the use of computer algebra but there remains a difficulty with the finiteness of the computations because all distributions except the normal have an infinite number of non-zero cumulants. One is led therefore to replacing finiteness of computations by ``finitely generated'' in the sense of recurrence relationships. In fact, it turns out that there is a natural definition in terms of the exponential model which is that the first and second derivative of the cumulant generating function, \(K\), lie on a polynomial variety. This generalises recent polynomial conditions on variance functions. This is satisfied by many examples and has applications to, for example, exact expressions for variance functions and saddle-point approximations.
Edgeworth expansions, cumulants, Computational problems in statistics, Statistical distribution theory, generalized linear models, saddle point approximations, computer algebra, exponential models
Edgeworth expansions, cumulants, Computational problems in statistics, Statistical distribution theory, generalized linear models, saddle point approximations, computer algebra, exponential models
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 0 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
