Hi all,
A little over 30 degrees of freedom, and a unit sample variance will do: a
t-distribution with these parameters will be a good proxy for the true Gaussian
distribution.
Regards,
"Paul" Moon Sub Choi.
?? Kentaro Fukumoto <fukumoto(a)dg8.so-net.ne.jp>jp>:
Hi,
what values of nu and s^2 should make our
distribution look like a
t-distribution of n-degrees of freedom?
Large nu, say 100, and s^2=1 do,
though I don't know why.
Hope it helps.
Kentaro
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l