Forums › Ask ACCA Tutor Forums › Ask the Tutor ACCA AFM Exams › Sensitivity Analysis
- This topic has 1 reply, 2 voices, and was last updated 9 years ago by John Moffat.
- AuthorPosts
- April 15, 2015 at 10:32 pm #241455
Hello Mr Moffat,
Suppose we are given a project lets say with the following appraisal (extract from June 2010 Seal Island)
Yr Cash Flows ($m) DF (10%) Present value
12 Construction (300) 0.909 (272·7)
13 Construction (600) 0.826 (495·9)
14 Construction (100) 0.751 (75·1)
15–44 operating surplus 100 10.59661 1059.7
44 Decommissioning (2189) 0.043 (94·3)NPV 122·1
My apologies the table format didn’t appear standard as it should be, however this link for the specific answer page might help
https://www.accaglobal.com/content/dam/acca/global/pdf/p4_2010_jun_ans.pdf
Sir i just wanted to clarify with the part of sensitivity analysis, can i calculate the sensitivity of Construction cost by
122.1/ (272.7+ 495.9 + 75.1) = 14.49% ( almost same with the answer given by the ACCA marking scheme but different workings)
However suppose in the scenario we are told to show by how much should it increase in the respective years of 2012, 2013, 2014 so that the overall NPV would be zero, how should be the workings?
And also sir, kindly can you explain what simulation means such as Monte Carlo simulation
Thanks once again for everything, I am really sorry to ask you too many questions.
Soud Said.
April 16, 2015 at 7:41 am #241475Your way of calculating the sensitivity is fine (and is the way that I would have done it).
With regard to how much it should increase in the respective years, I am not sure what you mean. You can only be expected to calculate sensitivity for one fact at a time. It would need each of the parts of the construction cost all to increase by 14.49%. The only alternative that could have been asked would be asking for the sensitivity of just one of the construction cost payments (e.g. just the payable in 2012) in which case you would take the same approach.
With regard to Monte Carlo simulation, you could not be asked to do any numbers – just to explain. What it would be is effectively setting up a spreadsheet showing all possible values of the items that are uncertain, calculating all the possible NPV’s that could result, and calculating the probabilities for each of these NPV’s. Then we would be in a position to estimate the probability of the NPV ending up in certain ranges (e.g. the probability of the NPV being great than zero, or the probability of the NPV being great than 100K and so on).
- AuthorPosts
- You must be logged in to reply to this topic.