Quick question about blended benchmarks. I'm hoping someone can help out.Say I have a blended benchmark that is made up of 60% Russell 1000 Growth and 40% Russell 1000 Value. Given the quarterly returns for the individual benchmarks, how could I calculate the annualized return for the blended benchmark? I know that I would need to blend the returns based on weight (60% and 40%), but I'm not sure at what point to blend the returns (i.e. do I blend the returns first and then calculate annualized return, or should I calculate annualized return for the two return series and then blend the results?).Thanks,Rob

aggregate the two indices first to construct your benchmark, then annualize the return.

Thanks for the quick reply, janickg.Do you know of any good articles or books that explain why it should be done this way?Thanks,Rob

QuoteOriginally posted by: janickgaggregate the two indices first to construct your benchmark, then annualize the return.This way you get a 60/40 benchmark with quartertly rebalancing. If you want annual rebalancing it gets it bit more complicated.

Well I guess one can construct an annual rebalancing by compounding the quarterly returns into annual segments then aggregating them. Or as I said, aggregate them quarterly then just annualize (1+r_qtr)^4 - 1

Thanks guys. I have a couple of follow-up questions:1. Is there a web site, book, or paper I can read to learn more about quarterly rebalancing vs. annual rebalancing of blended benchmarks?2. I notice that, if I calculate the annualized return using the two different methods that you mentioned (quarterly rebalancing and annual rebalancing), I get two different answers. Which one is right? Are they both right (under different contexts)?Thanks,Rob

If you take the average of the quarterly rebalancing, this is your quarterly periodic return - and when you annualize this quantity you are also making a rough assumption that this figure will happen four times a year at the same quantity.Of course when you actually compound and then aggregate annual figures, those are your true annual returns because you don't have to annualize these numbers under the assumptions above. You've already observed the periodic returns over the historical year and you're just calculating the return based on that.Not sure about literature out there myself, just in practice. But I would be interesting if anyone else has some interesting insights.I tend to build a simple case of two assets and weight them and ask myself, if I put in a dollar today, what do I expect intuitively to be my return on an annual figure the next year? And test the different methods of aggregating and see which one makes more sense.

QuoteOriginally posted by: rtougherThanks guys. I have a couple of follow-up questions:1. Is there a web site, book, or paper I can read to learn more about quarterly rebalancing vs. annual rebalancing of blended benchmarks?2. I notice that, if I calculate the annualized return using the two different methods that you mentioned (quarterly rebalancing and annual rebalancing), I get two different answers. Which one is right? Are they both right (under different contexts)?Thanks,Rob#2