# 99% confidence interval

Why does CFAI estimate this as ±3 Std dev from the mean? The value actually should be 2.58… I can understand rounding the 95% CI to 2std dev (1.96 actual)…But rounding 2.58 to 3 seems a bit much…

yeah, I noticed that too. I think they’re just being sloppy.

same here, i can’t ever figure whether to use 2.58 or 3…on the exam should we use 68% ± 1 std dev 90% ± 1.65 95% ± 1.96 99% ± 2.58 or the rounded figures??

use 2.58 if u find the answer then mark that, else go to the 3 option. I am pretty sure they will not mess up by giving both answers. If they do, then select the 2.58 because that is the more statistically correct. Similarly for the 95% interval too, they use 1.96 or 2, interchangably I would think. CP

I think, without looking at the z table, that 3 std’s away will probably still give you a 99% to 2 digits… and the fact that you can’t round up to 100% anyway… but I doubt they will give ambiguous answers… That’s prob’ly because usually people remember 68%, 95%, 99% for 1, 2, 3 sd instead of 1.65, 1.96, 2.58 for 90%, 95% and 99%…

well to be honest…I think thats pretty lame considering the amount of other stuff they expect you to remember. Or it may be the schweser training in me…where you have to calculate some things out to the 6th decimal place to get the correct answer.