Anyone ever measure their R-squared?

[oddballstocks]

I'm helping my company switch out some mutual funds for the 401k and one metric I always like considering is the R-squared.  It's not all that surprising but many of these funds are close index-huggers. 

 

So this got me thinking has anyone ever calculated the R-squared for their own portfolio?  Just based on discussions from this board I'm guessing people are going to be low.  If anyone has done this in the past I'd be interested in hearing about the experience, and if you're comfortable the R-squared valued.  It doesn't even have to be current.

 

[tombgrt]

I believe James Montier had a great paper about mutual funds and their R-Squared. If anyone is interested, I'll try to look it up.

[twacowfca]

Are you asking about the Coefficient of Correlation of a particular portfolio with the index?

If so, ours is not highly correlated because we buy a number of odd securities whose workout values are uncorrelated to the market; plus our largest holding, LRE, has generally tended to rise when the market has fallen.

 

You may be chasing your own tail in this project because diversified funds will have a certain amount of variance, and the tendency is for that variance to regress to the mean.

 

You will be better off to hold your outlier underperformers because they will generally tend to perform relatively well after their period of underperformance.  :)

[oddballstocks]

Twacowfca, yes Coefficient of Correlation.  I'm not surprised yours is low, that was my suspicion for most here as well. 

 

I agree with you on the under performance for the most part, although it seems some of the funds we have that are going to be removed are just bad funds, frequent manager changes and managers who just aren't that good.  One of the reasons I want to show R-squared is to drive home the point that we need a few more index funds in the 401k.  If I'm going to basically match the market with my funds I'd rather pay .2% to do it instead of 1.3%

[twacowfca]

Twacowfca, yes Coefficient of Correlation.  I'm not surprised yours is low, that was my suspicion for most here as well. 

 

I agree with you on the under performance for the most part, although it seems some of the funds we have that are going to be removed are just bad funds, frequent manager changes and managers who just aren't that good.  One of the reasons I want to show R-squared is to drive home the point that we need a few more index funds in the 401k.  If I'm going to basically match the market with my funds I'd rather pay .2% to do it instead of 1.3%

 

That's the most economical way to go if your strategy is to match the market. You might want to consider the WisdomTree funds that rebalance by value weighting instead of market cap weighting and the Dimensional Funds that are a little more intelligent in their selection than most index funds.

[Packer16]

We did the same exercize earlier this year and were able to add some non index funds like Sequoia, Longleaf, Third Ave, Fairholme, Harboe Intl and the Wellington sector funds.  We had Vanguard as our administrator so we already had alot of index products.

 

Packer

[mhdousa]

Only when I was a kid and wasn't sure if it was big enough....