2

Where do you find, or how do you calculate, the tracking error for an index fund? Every guide I've read about index funds mentions tracking error (the difference between the fund's return and the benchmark its tracking) as an important metric, but stock/fund listing sites like Yahoo and elsewhere don't show it.

Is it typically not listed on websites, or am I misreading them?

To calculate it, would I simply take the last 1 year return percent, and subtract that from the benchmark's change?

0
5

Courtesy of Investopedia

https://www.investopedia.com/terms/t/trackingerror.asp

Tracking Error = Standard Deviation of (P - B)

For example

Note Compare is a scaled version of Pprices.

Pprices  = 1000, 1100, 1150, 1300, 1250, 1150, 1200
BmkIndex = 1050, 1150, 1200, 1350, 1300, 1200, 1150

Compare  = 1000/1.1, 1100/1.2, 1150/1.3, 1300/1.4, 1250/1.5, 1150/1.6, 1200/1.7

enter image description here

P = 0.1, 0.0455, 0.1304, -0.0385, -0.08, 0.0435
B = 0.0952, 0.0435, 0.125, -0.037, -0.0769, -0.0417
C = 0.0083, -0.035, 0.0497, -0.1026, -0.1375, -0.0179

enter image description here

Tracking Error of B = Standard Deviation of (P - B) = 3.43%
Tracking Error of C = Standard Deviation of (P - C) = 1.36%

Despite the wide divergence of Compare from Pprices its tracking error is smaller.

3

To calculate it, would I simply take the last 1 year return percent, and subtract that from the benchmark's change?

That would give you a single data point, which probably isn't very meaningful. Suppose there was no difference over the year, but a huge tracking error over the first 6 months, which was reversed in the second 6 months? To calculate a meaningful error rate, you need to calculate the standard deviation of absolute error using several data points to give you a more accurate value. Using weekly or even monthly data would give you many more data points and give you a more accurate tracking error.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.