What's new

Is there any general rule to keep decimal points (for probabilities, etc)


New Member
For example, in the binomial tree model, given the volatility, the u=exp(\sigma*\sqrt{\delta t}). In the following example, u=exp(0.22*\sqrt{0.5}). Moreover,
p=(exp((r-q)\delta t)-d)/(u-d)
I think I understand how to do the following problem. However, my question is, how many digits should we keep in the exam?

Like in the problem, there are many layers that need rounding, which makes the final answer (216.67) can be off, depending on how to round u,d, and thus (u-d) in p, and so on. For example, keeping 4 digits or 2 digits of u or d (and thus p), will make the results different.

I hope we can just find the closest answer in the exam? (I assume no two choices will be as close as/within rounding error difference? finger-crossed)


Binomial Tree.jpg
Last edited by a moderator:

David Harper CFA FRM

David Harper CFA FRM
Staff member
Hi @liyi1989 I format to 4 decimals, [2nd] [Format] 4 [Enter], however please note that if you store to memory and recall from memory--ie.., [STO] and [RCL]-- you don't need to worry about the precision (stored values are accurate). Styles vary but when i work some of these problems with the Texas Instruments BA II+, I very much rely on the STO/RCL. Thanks,