I'm just wondering if anyone could give me a bit of advice on this. This relates to CCEA's S1 exam questions.
$Z \sim \text{N}(0, 1)$
Let's say $\phi(z) = 0.5015$ Find z.
Here is an extract of the normal probability table that we're given in the exam.
Now, our teacher said that, 'as far as he knows', if you cannot solve $\phi^{-1}$ for an exact value, you should always choose the value in the (ADD) column that gives less than the actual value.
So, using his advice, I would choose $z = 0.003$, since that would give $\phi(0.003) = 0.5012$ rather than choosing $z = 0.004$ which would give $\phi(0.004) = 0.5016$.
I just have concerns with this method, because he didn't seem totally sure that this is what you do, and 0.5016 is obviously more accurate than 0.5012 as they give differences of $10^{-4}$ and $3\times 10^{-4}$ respectively. This might be the correct method, but I just need to confirm it (exam is tomorrow!).
Thanks for any help!
Kelsey
P.S. I apologise if this has already been answered; I've looked on different forums, including this one, and I can't find anything that relates to this question.