Does anyone remember the easier way to calculate the skewess of the aggregate claims distribution which is used to determine the parameter values given the formula 2/a^(1/2) - i.e translated gamma. I remember being shown this when this topic used to be part of 106 but unfortunately i have lost the notes i made for this subject.
I think you're referring to "skewness is 2alpha/deltacubed" Or perhpas coefficient of skewness is 2/rootalpha=skewness/sigmacubed Or even that 2deltaY has a chi-squared distribution with 2alpha df. Most of this stuff is on page 12 of the Tables (but not the first bit!)