Can anyone tell me when do we use continuity correction? I know its when we are using using Central limit theorem on discrete distributions but I saw many places where it wasn't done and few where it was done ? If there is something else, that we consider while considering continuity correction please help ! ASAP!
It is used when we approximate discrete probability into continuous.( Also vice versa but not generally used) Eg. When we have to approximate binomial to normal
The continuity correction works very well when you have exact symmetry and more generally works well when you're towards the mean of the binomial, and often does badly when you are in the far tail of asymmetric distributions (p far from 0.5), though it depends on which direction you're looking.
