I have drawn 100 sample of size 10 from standard normal distribution using r . after that for each repetition the 95% t-confidence intervals have been plotted on the same graph paper and those intervals that don't cover the empiracal population mean '0' have been marked in 'red'. But what's the interpretation of this kind of graphs ? 
Asked
Active
Viewed 13 times
0
saudade
- 165
- 4
-
How many 'red bars' would you expect if you compute a 95% confidence interval? – Jan 06 '16 at 19:46
-
less than 5 .. ! as the probability of 'red bans' is theoretically 0.05 . – saudade Jan 06 '16 at 19:49
-
my question is , here we already know the mean beforehand. then why are we doing this ..? – saudade Jan 06 '16 at 19:51