I have the following Question in Business Statistics:
For n>100, i learned, that we can use the standard normal
distribution instead of the t-distr.
Does this also mean that for hypothesis tests of µ (pop. mean) I have to perform a Z-test for samples>100 in the sigma-unknown case (with the respective labels like Zstat/Zcrit etc.) and not like usually (for sigma-unknown) a T-test?
Moreover, with 2 populations the hyp test of µ works with µ1-µ2.
Also here in the Sigma unknown case if n1+n2 (the sum of both
sample sizes) is >100, i should use a Z-Test rather than a
Am i right?