I'm trying to work out the sufficient statistic for a normal distribution. I've figured out the statistic for when mu is unknown as that's simply x-bar. But when I thought that just S^2 was the sufficient statistic for sigma^2, I was told that was incorrect. Additionally, when both are unknown it's not just the two sufficient statistics of the constituents like it is with a Gamma distribution.
I know I'm missing something here but I'm not sure what...
Advertise on Talk Stats