Hello, I have questions around few intermediate calculations of Q7, part iv) Referring to the solution in Examiner's report. 1. Why are we considering TWRR for tracking error and not MWRR? Is it simply because TWRR is a more efficient indicator of performance? 2. How is the annualized performance of benchmark calculated? I thought it should be (1+(-1%))*(1+4%)*(1+1%)-1 = 3.99% . How are we calculating 1.31%? Can we not use 3.99% here? 3. What is the formula for calculating tracking error in Method A and Method B? How is the standard deviation calculated here (as defined for tracking error)? 4. Method A- What is the formula for calculating information ratio here? Will be helpful if someone can answer these questions. Thank you in advance!
Hi, 1) Yes if we can use TWRR then we should. The TWRR equation naturally gives us a nice formula for calculating individual annual returns for each year which we need for the tracking error. The MWRR number is an annualised number but it applies on average across the three years. 2) The clue is in the name here. If we need an annual figure we have to take the three years return and turn them into an annual number by cube rooting (1+i)^3. 3) Under method A: Tracking error = sqrt(variance of relative returns / N-1). Variance of relative returns is the sum [ (Xi - Xbar)^2 ] where Xi represent relative return in year i and Xbar represents the average relative return. N-1 being used as the data here only represents a sample rather than the whole population. Under method B: Tracking error = sqrt((sum (relative performance squared)) / N-1) Method B is not the method we would expect to use or see in the exam but the examiners deemed it to be acceptable. 4) Under both methods the information ratio is calculated as relative return / tracking error. Hope this helps. Please let me know if you have any further questions. Joe
