From CFA EOC Reading 11, question #8
“Willco is a manufacturer in a mature cyclical industry. During the most recent industry cycle, its net income averaged $30 million per year with a SD of $10 million (n=6 observations). Mngmt claims that Willco’s performance during the most recent cycle results from new approaches and that we can dismiss profitability expectations based on its average or normalized earnings of $24 million per year in prior cycles.”
I’m having trouble discerning whether to use a 1-tail test or 2-tail test here when finding the rejection point(s). I know the answer is 1-tail, and I can reason as to why (ie. management thinks this $30 mill is greater than normal, etc.). But to me it wasn’t so obvious, and more importantly, theoritically, I’m having trouble understanding the difference between the two.
I mean who’s to say I can’t set up a 2-tail test with +/- test stat and see whether it falls outside my interval and use a .025 level when looking up the 2-tail value. Would it still not be at the 0.05 sig level but spread out over left and right tail?? Yet the rejection point on right side would obviously be different. Blah…