Goalseek in paper A

[ss_98]

According to the Acted ASET solutions below (CM2 Sep 2022 Q1), we shouldn't be using goalseek in paper A, just wanted to check why this is, especially if the question doesn't specify interpolation etc.

" (1008 - P)^0.01 + 9999(8 - P)^0.01 = 10,201.12

This equation can’t be solved analytically (ie it’s impossible to make P the subject) so it could be very tempting to use Goal Seek in Excel instead. However, whilst that would be acceptable in CM2 Paper B, a more manual mathematical approach is required here. Fortunately, the question asks for the correct price to only the nearest dollar, so a trial-and-error approach is easy to implement.
Examiners confirmed that full marks would be awarded just for identifying the equation above without proceeding to find a numerical value for P."

 

[Adam S]

I guess because goal seek doesn't always actually give the exact answer - just the way it works it may only give an answer close to the truth due to the way it converges to the answer.
I've had it a few times in situations like this where there is a correct answer, say 7/12, but it gives 0.58 and then a bunch of random digits (I think as close as it's convergence algorithm can get to, given the formula).

But to be honest, in an exam I'd potentially just use goal seek anyway if you are stuck, knowing at most you'd likely only loose 1 or 2 marks.

 

[John Potter]

Goal seek is a great tool to use in Paper A as well as Paper B.

Extremely important - ALWAYS be honest about what you do in your exam. If you use Goal seek, it's fine just to say so and the worst that can then happen is that you don't get the marks for that question because you were meant to be doing something else. If you use Goal seek and pretend that you didn't, this could result in an immediate fail for lack of integrity.

But let's try to apply a bit of common sense here. Imagine we have an implied volatility question (Black-Scholes) for 2 marks. I would smash out a BS calculator in Excel, use Goal seek and say "Using goal seek in Excel, vol = 23%"
Instead, imagine the same question (Black-Scholes) for 5 marks with "you must show all your working". It's clear more is required here, surely? I would smash out a BS calculator in Excel, use Goal seek and see that the answer is vol = 23%.
Then I would "try 20%" and copy all my answers from Excel into Word.
I would write "This is too low, so let's try a higher volatility"
I would "try 30%" and copy all my answers from Excel into Word.
Finally, I would linearly interpolate to get the answer - and I would do the linear interpolation. I would NOT quote the Goal-seeked number at any point in my solution. So my linear interpolation answer might be 23.12% whilst anyone goal-seeking gets 23.17%, say, and might lose marks in the different ways discussed above.

Hopefully, we can apply similar common sense to other questions too - the number of marks available and the wording of the question should help us decide. If in doubt, do something sensible and be honest about what you've done.

Hey, maybe I'm going OTT with all of this? There's at least one parallel universe out there where students follow my above advice in a 5-mark implied volatility question and the ER says something like "Stronger candidates used goal seek whilst weaker candidates wasted time with linear interpolation."

John