>>11873567Sorry was away. I myself also have a stat degree, but this person of more of an ML stat person whilst I was more of a stochastic stat person. But what I mean was that theoretical knowledge as certified by a uni does not mean understanding.
>>11873599It does, but this is not what I was talking about. With extremel fat tailed distributions, low sample sizes can produce significantly lower p values than what is actually true. In fact, it is possible that you can move up in sample size whilst dropping p-value to then move up again and have it shoot up. Thats why replication is so important, and you should never believe any study that has not been replicated.
