You may notice that the returns for a Morningstar Category average listed in Research View (for example, Large Value) are different from the returns you get when you create a graph for the same category for the same time period. To illustrate the difference between the category averages in Research View and the category averages in the graphs of The Workstation, let's consider the three-month average of three funds in Category X with the following three-month return histories:
|
Category X |
Month 1 |
Month 2 |
Month 3 |
|
Fund A |
--- |
6.5 |
3.0 |
|
Fund B |
5.4 |
2.6 |
-1.5 |
|
Fund C |
4.6 |
1.0 |
5.0 |
A. Research View category averages:
1) Compound three-month return of each fund,
Fund A: none (it's not three months old)
Fund B: 6.5
Fund C: 10.93
2) then average them:
(6.5 + 10.93)/2 = 8.72
B. Graph category averages
1) Average each month's returns,
Month 1: (5.4 + 4.6)/2 = 5.0
Month 2: (6.5 + 2.6 + 1.0)/3 = 3.37
Month 3: (3.0 + -1.5 + 5.0)/3 = 2.17
2) then compound:
The formula for compounding is:
[(1+ M1/100)(1+M2/100)(1+M3/100)-1] x 100,
so the compounded average =
[(1+ 5.0/100)(1+3.37/100)(1+2.17/100) - 1] x 100 = 10.89
For the category averages in Research View, we take the average of the funds that exist for the time period and divide by the number of funds that exist for that time period. So in example A, we calculate the average using only Fund B and Fund C, because Fund A does not have a three-month return.
For the graph category averages in example B, we calculate the average monthly returns first and THEN compound, so we include month 2 and month 3 of Fund A.
This can cause a big difference in total returns, depending on the size of the returns that have been left out of the average. Also, example B results in monthly rebalancing, as if you took your money out of the funds at the end of each month, then redistributed it equally among the funds the next month. So it implies initial weights of ½, ½ at the start of the first months, and 1/3,1/3,1/3 at the starts of the second and third months, In contrast, Example A assumes initial weights of ½, ½ being held for all three months—you let your winners grow with method A and your losers sink. "