not going to lie i still don't get why we've double standardised for Q7bi. great explanation for everything else but it's still a bit unclear to me logically what's going on. maybe i lack a bit of inherent understanding but why are you trying to minimise the probability of accepting Ho?
A few people commented and sent emails about this one, hopefully this clears it up a little ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-AYxqrNJGMaY.html
I don't think I explained this one very well. I will try to get back to you if I have anymore insight. I should have mentioned the only reason I did it like that is because the alternative is setting up the same expression as < -1.6449, which leads to an answer like n
@@rtwodrew2 Is there any way you could do another explanation of questions like this, maybe going through a couple similar - it's something I really struggle with and don't quite understand even after this video?
@@rtwodrew2 Thank you very much, esp. given you seem to have been the only one to cover this paper on youtube. This + the TMUA videos (even if I didn't get a place) have been a godsend, much much appreciated!
Hi! I don't know if you still need help or not, but basically if in the original expression we have P(Xx) and we standardise it, the z-values will be positive
We're not estimating the binomial, just thinking about the mean of the sample, which is normally distributed by the CLT. It doesn't matter what the original distribution was. There's actually no continuity correction in FS1 anywhere