[Deleted]

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
bogledogle87
Posts: 231
Joined: Wed Sep 26, 2018 7:03 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by bogledogle87 » Thu Jul 11, 2019 2:27 pm

lazyday wrote:
Thu Jul 11, 2019 12:39 pm
Investing without following an index seems to offer significant benefits for a fund like this. Of course, the downside is that when the fund does poorly, it can be hard to know why. And if the fund isn’t index hugging, then there should be times where it underperforms.
I definitely agree. I now realize I grossly overestimated my own comfort level with this part of it.
VTWAX and chill

Topic Author
hdas
Posts: 1320
Joined: Thu Jun 11, 2015 8:24 am

[Deleted]

Post by hdas » Wed Jul 17, 2019 9:29 am

[Deleted]
Last edited by hdas on Wed Jan 29, 2020 8:04 pm, edited 1 time in total.
....

muffins14
Posts: 249
Joined: Wed Oct 26, 2016 4:14 am

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by muffins14 » Wed Jul 17, 2019 10:54 am

Since this is an equity-only fund, it would be more appropriate to only use the equity factors (and exclude Term and Credit)

Re-running the regression there and using daily returns, one finds that USMF has statistically significant loadings on market, size, and BAB. It does result in positive alpha, but it's also not statistically significant

Topic Author
hdas
Posts: 1320
Joined: Thu Jun 11, 2015 8:24 am

[Deleted]

Post by hdas » Wed Jul 17, 2019 11:11 am

[Deleted]
Last edited by hdas on Wed Jan 29, 2020 8:04 pm, edited 1 time in total.
....

muffins14
Posts: 249
Joined: Wed Oct 26, 2016 4:14 am

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by muffins14 » Wed Jul 17, 2019 11:28 am

No resource at my fingertips now, but I believe credit is to measure the premium of lower-credit bonds over higher-credit bonds, and term is to measure the premium of longer-term bonds over shorter-term bonds. In the case of a portfolio without bonds I don't think it would make sense to include those variables. But again I'm not an expert here, maybe someone else would chime in.

As an aside -- of course you can always include as many variables in any type regression as you want, but you'll start to lose interpretability, robustness against correlated features, and the ability to make good inferences about the meaning of the coefficients.

Topic Author
hdas
Posts: 1320
Joined: Thu Jun 11, 2015 8:24 am

[Deleted]

Post by hdas » Wed Jul 17, 2019 11:45 am

[Deleted]
Last edited by hdas on Wed Jan 29, 2020 8:04 pm, edited 1 time in total.
....

User avatar
vineviz
Posts: 7236
Joined: Tue May 15, 2018 1:55 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by vineviz » Wed Jul 17, 2019 12:10 pm

hdas wrote:
Wed Jul 17, 2019 11:11 am
muffins14 wrote:
Wed Jul 17, 2019 10:54 am
Since this is an equity-only fund, it would be more appropriate to only use the equity factors (and exclude Term and Credit)
People keep repeating this but I haven't seen a good reason for it. Do you have a reference from a credible source. Thanks :greedy
The reasons don't have anything specifically do with whether the variable is an "equity factor" or a "bond factor", but anytime you see a Fama-French regression with high values for unadjusted R2 or regression F statistic combined with a large number of statistically insignificant variables then you have a big red flag for over-fitting.

Over-fitting results from using more (correlated) independent variables than are necessary to explain the variance in your data.

There are plenty of formal tests for this (many econometrics books should cover it) but I don't have a solid reference handy. Look for a source that describes over-fitting, collinearity, and/or variance inflation.

What is happening is usually that the correlated variables are insignificant individually but significant in combination (because they are correlated), increasing the raw coefficient of determination (i.e. R2). That explanatory power has to go somewhere, so it (falsely) goes to alpha.

A quick and dirty check is to drop one or two of the least significant variables and see if the adjusted R2 and/or regression F-stat go up. If the adjusted R2 goes up while the raw R2 goes down, for example, chances are good that you have compromised your DoF somehow.

In this case with USMF reducing the factor set to RMF, SMB, HML, & MOM increases the adjusted R2 from 92.4% to 93.8% and the F-stat from 34.29 to 84.31. As a broad rule of thumb, that's a good indication that the smaller variable set is maximizing both explanatory power and degrees of freedom.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

Topic Author
hdas
Posts: 1320
Joined: Thu Jun 11, 2015 8:24 am

[Deleted]

Post by hdas » Thu Jul 18, 2019 3:49 pm

[Deleted]
Last edited by hdas on Wed Jan 29, 2020 8:05 pm, edited 1 time in total.
....

User avatar
vineviz
Posts: 7236
Joined: Tue May 15, 2018 1:55 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by vineviz » Thu Jul 18, 2019 4:19 pm

hdas wrote:
Thu Jul 18, 2019 3:49 pm
1. In my portfolio of individual stocks, the situation is the reverse, I need to add more factors to increase the R^2, only Term and Alpha are significant.
It's important to distinguish between unadjusted R^2 (which will ALWAYS go up with the addition of more independent variables) and adjusted R^2 (which will increase only so long as the added variable adds enough explanatory power to compensate for the smaller number of degrees of freedom).

For instance, using a random portfolio of 50 stocks I plotted the adjusted R^2 and unadjusted R^2 with various numbers of factors.

Image

You can see that unadjusted R^2 increases no matter how many variables are added (within the rounding error reported by PortfolioVisualizer, anyway) whereas the adjusted R^2 peaks with a four-factor model.
hdas wrote:
Thu Jul 18, 2019 3:49 pm
2. A priori it seems to me that yield curve and credit spreads are more important (macro) factors than the style equity factors.
That wouldn't be my prior belief, but it does raise a hypothesis that can be tested. As it turns out, term and credit factor are NOT more powerful at explaining the cross-section of stock returns than so-called equity factors (e.g. market, size, value, quality, etc.)
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

Topic Author
hdas
Posts: 1320
Joined: Thu Jun 11, 2015 8:24 am

[Deleted]

Post by hdas » Thu Jul 18, 2019 5:03 pm

[Deleted]
Last edited by hdas on Wed Jan 29, 2020 8:05 pm, edited 1 time in total.
....

User avatar
vineviz
Posts: 7236
Joined: Tue May 15, 2018 1:55 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by vineviz » Thu Jul 18, 2019 6:07 pm

hdas wrote:
Thu Jul 18, 2019 5:03 pm
vineviz wrote:
Thu Jul 18, 2019 4:19 pm

It's important to distinguish between unadjusted R^2 (which will ALWAYS go up with the addition of more independent variables) and adjusted R^2 (which will increase only so long as the added variable adds enough explanatory power to compensate for the smaller number of degrees of freedom).
If you look at the link I sent, you'll notice that both the raw and adjusted R^2 go up as I add more factors. I start with 3 factors and keep adding one by one sequentially. Cheers :greedy
Without access to the holdings and regression settings I can't help you further diagnose the causes, but I can tell you that's a problematic regression.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

Topic Author
hdas
Posts: 1320
Joined: Thu Jun 11, 2015 8:24 am

[Deleted]

Post by hdas » Mon Jul 22, 2019 1:23 pm

[Deleted]
Last edited by hdas on Wed Jan 29, 2020 8:05 pm, edited 1 time in total.
....

User avatar
vineviz
Posts: 7236
Joined: Tue May 15, 2018 1:55 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by vineviz » Mon Jul 22, 2019 7:55 pm

hdas wrote:
Mon Jul 22, 2019 1:23 pm
vineviz wrote:
Thu Jul 18, 2019 6:07 pm
Without access to the holdings and regression settings I can't help you further diagnose the causes, but I can tell you that's a problematic regression.
I get the same effect when running the factor regression on my beta portfolio, here are the specs:

> Last 3 years
> VUG 30%, USMV 30%, IJS 20%, XSLV 20%
> You can start adding factors sequentially, start with AQR 3 factor, then 4 factor, then add quality, then BAB, finally Term + Credit.
> Notice how at every step you get higher adjusted R^2
For whatever reason, the regression on this portfolio (VUG/USMV/IJS/XSLV) is much more robust than whatever you were regressing in the other thread: R-squred in the 98% range instead of high 60s and low 70s, f-stat in the hundreds instead of the low-teens. T

Even so, the regression quality is marginally higher without including CDT (the adjusted R-squared stays at 98.2% but the f-stat is higher without than with). What makes the difference here, versus some other earlier examples, is that all 7 factors (once credit is excluded) are statistically significant.

With such a significant allocation to USMV and XSLV, I'm not surprised that TERM is statistically significant since the regression doesn't explicitly contain a low volatility or minimum variance factor: HML, MOM, and BAB are left to pick up the weight and they are too imperfect as proxies.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

jmk
Posts: 581
Joined: Tue Nov 01, 2011 7:48 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by jmk » Mon Jul 22, 2019 8:34 pm

vineviz wrote:
Thu Jul 18, 2019 4:19 pm
It's important to distinguish between unadjusted R^2 (which will ALWAYS go up with the addition of more independent variables) and adjusted R^2 (which will increase only so long as the added variable adds enough explanatory power to compensate for the smaller number of degrees of freedom).
What does adjusted r2 mean?

Topic Author
hdas
Posts: 1320
Joined: Thu Jun 11, 2015 8:24 am

[Deleted]

Post by hdas » Mon Jul 22, 2019 8:52 pm

[Deleted]
Last edited by hdas on Wed Jan 29, 2020 8:05 pm, edited 1 time in total.
....

Day9
Posts: 880
Joined: Mon Jun 11, 2012 6:22 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by Day9 » Fri Aug 02, 2019 4:02 pm

Bumping this thread to report that just released today Larry Swedroe gave his opinion on the Vanguard Multifactor fund on the Bogleheads Podcast hosted by Rick Ferri.

Timestamp 35:20
https://bogleheads.podbean.com/e/episod ... ick-ferri/

Summary of Larry's answer: Anything Vanguard does is likely to be done very well. They are highly ethical and have good researchers. Larry's firm took a look at these funds. They are well designed and cheap, living up to Vanguard's reputation. So this is worth looking into if you are looking into diversifying away from a pure market cap index portfolio. The only negative, if you can call it that, is they do not offer very deep exposure. Bogleheads sometimes focus too much on expense ratios when instead sometimes you can get much deeper factor exposure for a little more cost.
I'm just a fan of the person I got my user name from

HippoSir
Posts: 111
Joined: Tue Jul 03, 2018 2:56 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by HippoSir » Fri Aug 02, 2019 4:31 pm

Day9 wrote:
Fri Aug 02, 2019 4:02 pm
They are well designed and cheap, living up to Vanguard's reputation. So this is worth looking into if you are looking into diversifying away from a pure market cap index portfolio. The only negative, if you can call it that, is they do not offer very deep exposure.
I'm curious what funds he's comparing them to and how he's determining exposure. VFVA in particular seems to have very deep value exposure compared to most comparable funds that are easily available (even compared to DFA funds!), through the lens of a https://www.portfoliovisualizer.com/ factor regression.

Day9
Posts: 880
Joined: Mon Jun 11, 2012 6:22 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by Day9 » Fri Aug 02, 2019 4:43 pm

HippoSir wrote:
Fri Aug 02, 2019 4:31 pm
Day9 wrote:
Fri Aug 02, 2019 4:02 pm
They are well designed and cheap, living up to Vanguard's reputation. So this is worth looking into if you are looking into diversifying away from a pure market cap index portfolio. The only negative, if you can call it that, is they do not offer very deep exposure.
I'm curious what funds he's comparing them to and how he's determining exposure. VFVA in particular seems to have very deep value exposure compared to most comparable funds that are easily available (even compared to DFA funds!), through the lens of a https://www.portfoliovisualizer.com/ factor regression.
If you listen to the whole interview I linked he said that his firm dropped DFA small value fund because even that did not offer deep enough exposure! A Bridgeway fund (probably BOSVX) offered deeper exposure, but at higher cost. But the much deeper exposure was worth the slightly higher cost. In the interview he mentions the funds he invests in, and his book Your Complete Guide to Factor-Based Investing has an appendix at the end where he recommends specific funds and ETFs.
I'm just a fan of the person I got my user name from

Helot
Posts: 264
Joined: Sat Mar 03, 2007 3:05 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by Helot » Fri Aug 02, 2019 6:17 pm

HippoSir wrote:
Fri Aug 02, 2019 4:31 pm
VFVA in particular seems to have very deep value exposure compared to most comparable funds that are easily available (even compared to DFA funds!), through the lens of a https://www.portfoliovisualizer.com/ factor regression.
My exact reaction upon listening to this interview during my drive today. I was waiting to return home to review the factor regressions on Portfolio Visualizer.

Day9
Posts: 880
Joined: Mon Jun 11, 2012 6:22 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by Day9 » Fri Aug 02, 2019 6:58 pm

Helot wrote:
Fri Aug 02, 2019 6:17 pm
HippoSir wrote:
Fri Aug 02, 2019 4:31 pm
VFVA in particular seems to have very deep value exposure compared to most comparable funds that are easily available (even compared to DFA funds!), through the lens of a https://www.portfoliovisualizer.com/ factor regression.
My exact reaction upon listening to this interview during my drive today. I was waiting to return home to review the factor regressions on Portfolio Visualizer.
You guys are right however portfoliovisualizer shows -0.28 loading on Momentum (when set to AQR 4-factor model). DFA Small Value and Bridgeway BOSVX do not have that problem (if you consider that a problem).
I'm just a fan of the person I got my user name from

Helot
Posts: 264
Joined: Sat Mar 03, 2007 3:05 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by Helot » Fri Aug 02, 2019 7:11 pm

I consider VFMFX (0.18 ER) the Vanguard equivalent of DFA Vector Equity (0.32 ER + Access Cost of UNK)

Using AQR 4 Factor with Quality:

Image

And see the similarity when the two funds are compared over a common time period:

Image

I hope Vanguard releases a similar ex-US option in the future making a Vanguard global multifactor 2 fund equity portfolio a well priced and accessible option. I’d combine these funds with the TSP G Fund in retirement and be done with all this portfolio perfecting nonsense. 8-). That is, until I get the ole’ slice and dice itch again and consider adding REIT and commodities funds ... :oops:

Final note, the construction of of the Vanguard fund should be no surprise as the portfolio manager is a former student of Eugene Fama as well as a DFA alum ...

Image

klaus14
Posts: 476
Joined: Sun Nov 25, 2018 7:43 pm

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by klaus14 » Sat Nov 30, 2019 2:39 pm

Day9 wrote:
Fri Aug 02, 2019 4:43 pm
HippoSir wrote:
Fri Aug 02, 2019 4:31 pm
Day9 wrote:
Fri Aug 02, 2019 4:02 pm
They are well designed and cheap, living up to Vanguard's reputation. So this is worth looking into if you are looking into diversifying away from a pure market cap index portfolio. The only negative, if you can call it that, is they do not offer very deep exposure.
I'm curious what funds he's comparing them to and how he's determining exposure. VFVA in particular seems to have very deep value exposure compared to most comparable funds that are easily available (even compared to DFA funds!), through the lens of a https://www.portfoliovisualizer.com/ factor regression.
If you listen to the whole interview I linked he said that his firm dropped DFA small value fund because even that did not offer deep enough exposure! A Bridgeway fund (probably BOSVX) offered deeper exposure, but at higher cost. But the much deeper exposure was worth the slightly higher cost. In the interview he mentions the funds he invests in, and his book Your Complete Guide to Factor-Based Investing has an appendix at the end where he recommends specific funds and ETFs.
I am glad VFMF value exposure was not THAT deep. CAGR Since inception:

VFMF 3.67%
BOSVX -4.03%
DFSVX -0.17%
VFVA 1.38%

I guess MOM exposure and not being only small meant better performance for VFMF.
SP500 CAGR in the same period was 10.75% :shock: . Compared to that, they all suck.
Last edited by klaus14 on Sat Nov 30, 2019 4:23 pm, edited 1 time in total.
35% US, 20 ExUS Dev, 10% EM, 10% EM Bonds, 10% Gold, 10% EDV, 5% I/EE Bonds. 50% value tilt in stocks.

Northern Flicker
Posts: 6038
Joined: Fri Apr 10, 2015 12:29 am

Re: VFMF (Vanguard Multifactor) == Closet Value Index Fund

Post by Northern Flicker » Sat Nov 30, 2019 4:19 pm

Thanks. I don't see how anybody thinks that the yield curve and credit spreads are not extremely relevant factors.
With respect to equity returns, the yield curve and credit spreads are macroeconomic factors. Including them would be like including the a published inflation rate. In the 1980’s there were factor models that incorporated a large number of macro-economic factors to try to do tactical asset allocation. These worked well at times (strong allocations to fixed income in 1987 leading up to the Oct 1987 crash) but ultimately were a failure.

Equity factor models extend CAPM with a focus on actual properties of individual stocks that are believed to explain their return. They attempt to explain CAPM alpha in a systematic way. Macro-economic factors would be absorbed primarily into market beta return with residual effects incorporated into alpha in an equity factor model. Increases in interest rates that lead to a higher discount rate for discounting future cash flows back to a present value would apply to the whole market, hence the absorption into market beta as a primary effect.

I think there is a weak case to be made for including term, in that some stocks, especially ones with robust, stable dividends, do incorporate some independent term exposure. But the effect of interest rates on equities is broad due to the effect of changes to the discount rate for future cash flows and it is unclear whether it matters whether you discount dividend flows or revenue flows back to a present value. The independent term exposure also has substantial overlap with BaB.

Factor regressions are useful but also have limits in their generalizability due to strong sample bias in the samples used for the regressions.
Risk is not a guarantor of return.

Post Reply