Fama and French: The Five-Factor Model Revisited
Re: Fama and French: The Five-Factor Model Revisited
deleted duplicate post.
Last edited by Apathizer on Tue Jan 25, 2022 10:12 pm, edited 2 times in total.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
Re: Fama and French: The Five-Factor Model Revisited
Small growth is more likely to underperform so there's no point in tilting towards it.
I didn't specify this but I think you understood and are just being contrarian as others have pointed out. We have significant evidence that certain sectors of the market are likely to outperform TSM. We also have significant evidence that other sectors are likely to underperform the market like small growth.
So the point isn't diversification for its own sake, but tilting towards segments of the market we have evidence will likely outperform the total market. That's why it makes sense to tilt toward those factors.
Of course we can't be absolutely certain those factors will outperform in the future, but we have theoretical and empirical evidence they will. Either you don't understand or just being contrarian. A cap weight Total market index fund diversifies away any and all factors except the market. If that's how you're inclined to invest that's fine. But there's evidence that tilting toward segments of the market that are likely to outperform will improve overall returns.
Last edited by Apathizer on Tue Jan 25, 2022 9:01 pm, edited 1 time in total.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
Re: Fama and French: The Five-Factor Model Revisited
Stocks are riskier than bonds because their cash flows are much more volatile. A bond gives you a predictable cash flow, if you buy a 1,000 bond with a 6% coupon, you will get $60 a year in cash flow year after year until the bond matures. With a bond, you get your principal back when the bond matures. There is a risk of default, if the company goes bust the bond holders are among the first in line to get paid back when the assets of the company are liquidated.vineviz wrote: ↑Sun Jan 23, 2022 4:33 pmI guarantee you that 99% of investors can't articulate why stocks are riskier than bonds any more clearly than academics have articulated why the other factors exist. Heck, 99% of investors can't even properly identify the risk-free asset, much less explain what an equity risk premium is or how to calculate it.
"I don't understand it therefore it can't exist" is a pretty weak argument, if you ask me.
A stock generates cash, that is if it generates a positive cash flow after expenses. Hopefully a stock would generate earnings which is a positive cash flow after expenses and certain non-cash charges like depreciation. No guarantee a stock will generate positive cash flows or earnings. No guarantee on the consistency of the cash flow and earnings. Earnings can be volatile. If the company goes bust, the holders of stock are among the very last in line to get paid back when the assets of the company are liquidated.
Bonds trade mostly on the level of interest rates and credit quality. Falling interest rates will boost the price of a bond and rising interest rates will be a drag on the price of the bond. As the bond gets closer and closer to maturity, the bond will trade closer and closer to its par value, that is the amount that bond issuers promise to repay the bond holders at maturity. Credit quality also affects bond prices, the higher the credit quality, the better.
Stocks trade at a multiple of earnings, the higher the earnings growth rate and the greater the consistency of the earnings growth, the more the market is willing to pay for a dollar of earnings. Stocks are riskier because of investor emotion, enthusiasm for stocks drives up earnings multiples and investor indifference for stocks tends to dampen earnings multiples. Stocks are more subject to the speculative fervor and investor emotion than bonds are. Stock prices also take into account the strength of the balance sheet.
Another reason that stocks are riskier is that stock analysts tend to be more optimistic in nature than bond analysts. The stock analyst sees opportunities where the bond analyst sees problems. Since stocks have higher expectations built into them than bonds, stocks are more likely to disappoint, low expectations are easier to beat than low expectations. The higher potential to disappoint also leads to higher volatility.
A fool and his money are good for business.
-
- Posts: 15369
- Joined: Fri Apr 10, 2015 12:29 am
Re: Fama and French: The Five-Factor Model Revisited
Part of the risk premium of equities is compensation for the uncertainty that risk is priced accurately. Equity risk is constantly repricing to price in new information. That's why equities are volatile and risky.km91 wrote: ↑Tue Jan 25, 2022 6:39 pmI work in credit, which explains why I find the factors compelling yet baffling. We need to explain every risk, how else could we set the price?Northern Flicker wrote: ↑Tue Jan 25, 2022 4:40 pm
You can only demand an actual higher return with fixed income instruments.
If equity risk were as well understood as credit risk, equities would be less volatile, and their prices would rise to eliminate a substantial part of their risk premium, reflective of less uncertainty. If the uncertainty associated with small cap stocks and/or value stocks and/or other factors are not risks you wish to take, that's great. Factor and asset pricing theory (which includes CAPM) have demonstrated that you can diversify those risks away by holding the market portfolio.
But I don't believe for a second that equity market risk can be modeled in a fundamental way at the level of detail and understanding at which credit risk is modeled.
Re: Fama and French: The Five-Factor Model Revisited
Exactly. I changed my post to reflect this. I should have originally written likely positive factors. Yes, diversifying across all factors is essentially what a TSM index is.muffins14 wrote: ↑Tue Jan 25, 2022 8:12 amYou are making exactly the opposite point of the poster you are quoting, and I think you know that.rkhusky wrote: ↑Tue Jan 25, 2022 7:38 amExactly. Sometimes Small Growth will outperform. Sometimes Small Value will outperform. Sometimes Large Growth will outperform. Sometimes Large Value will outperform. And the correlations between them range from 0.61 - 0.89. So, diversify by investing in all of them with TSM.
If you really want to tilt to one part of the market, you don't want to invest in all the factors, because they will cancel out and you will be left with TSM. And most people don't invest in factors anyway, they invest in half-factors, i.e. the long portion of the factor.
The correlation with the market and the return of various market segments over the last 20 years is below:
(note that the annualized return for TSM was 9.78%):
1/1/2002 - 12/31/2021
(correlation with TSM, annualized return)
LCG - 0.95, 10.71%
LCV - 0.95, 8.37%
SCG - 0.92, 11.00%
SCV - 0.91, 9.58%
If you weren't invested in growth during this period, you missed out on a lot of return, which will take years to make up. Best not to make bets on one part of the market, and just invest in it all.
Firstly, you know well that the TSM is not "all the factors", it is one of them, the market factor. This is by definition, like "how many ounces is a gallon".
Secondly, Growth and Value are not two factors that cancel each other out in the TSM. There is one factor, HmL, or value minus growth. The TSM has zero because it has equal parts growth and value, by definition.
Thirdly, this poster is recommending to diversify across all factors, which explicitly means NOT investing in only the TSM. Correspondingly it means not investing in equal parts growth and value, nor following market cap for small and growth. The poster is implying to have non-zero loadings on the factors in the portfolio, which means more value than growth, and more small compared to the TSM. Factor diversification implies non-zero (and positive) allocations to the factors, not zero allocations to the factors.
I think you know all of this, and are just using their quote and stating the exact opposite of their original intention as some tongue-in-cheek humor or something.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
-
- Posts: 15369
- Joined: Fri Apr 10, 2015 12:29 am
Re: Fama and French: The Five-Factor Model Revisited
No, it most certainly is not that. TSM diversifies away (ie eliminates) all factor exposures other than to the market factor.Apathizer wrote: Yes, diversifying across all factors is essentially what a TSM index is.
Re: Fama and French: The Five-Factor Model Revisited
Based on what? Returns from 50 years ago? What evidence is there that that data is relevant today? There are many things different in today's market - high speed trading, cloud computing, machine learning, apps like RobinHood, the ease of day trading.
Of course, but determining which sectors are likely to outperform is uncertain. Over the last 20 years, SCG has outperformed SCV. Perhaps things have changed and SCG will outperform for another 20 years. Or perhaps SCG has been bid up too high and LCV will take its place for the next 20 years. Perhaps SCV will have a burst of outperformance, but will it be enough to overcome the preceding 20 years of underperformance? No one knows.
Re: Fama and French: The Five-Factor Model Revisited
But people don't invest in the factors. They invest in market segments. And TSM contains all the US market segments.Northern Flicker wrote: ↑Tue Jan 25, 2022 10:52 pmNo, it most certainly is not that. TSM diversifies away (ie eliminates) all factor exposures other than to the market factor.Apathizer wrote: Yes, diversifying across all factors is essentially what a TSM index is.
Agreed though that TSM does not contain factor exposures. And market segments SCV, SCG, LCV, LCG individually do contain factor exposures.
Re: Fama and French: The Five-Factor Model Revisited
No, return data for at least the last 40 years. Robinhood and all the other things that facilitate rapid trading don't change this fact.
Agreed, all we can do is look at the best evidence and analyze it. That shows theoretical and empirical reason to expect SCV to out-perform since many investors pay unreasonably high price for growth.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
-
- Posts: 15369
- Joined: Fri Apr 10, 2015 12:29 am
Re: Fama and French: The Five-Factor Model Revisited
You get the market return from the market portfolio. There are no other factor exposures. All individual stock alpha under CAPM is diversified away. Factors other than the market factor (which try to model what is going on in alpha under CAPM) are diversified away in the market portfolio.rkhusky wrote: ↑Wed Jan 26, 2022 7:59 amBut people don't invest in the factors. They invest in market segments. And TSM contains all the US market segments.Northern Flicker wrote: ↑Tue Jan 25, 2022 10:52 pmNo, it most certainly is not that. TSM diversifies away (ie eliminates) all factor exposures other than to the market factor.Apathizer wrote: Yes, diversifying across all factors is essentially what a TSM index is.
Agreed though that TSM does not contain factor exposures. And market segments SCV, SCG, LCV, LCG individually do contain factor exposures.
Factor exposures are properties of individual stocks, and properties of portfolios. They are measures of risk. When you combine different portfolios into a single portfolio, risk is not additive-- it is not just aggregated from all of the risks of components. A portfolio of about 20% total stock market and 80% treasury bonds is less risky than either 100% bonds or 100% stocks. If you retained the risk of every individual stock and every factor exposure of every stock and/or stock segment in the market portfolio, it would be so risky that you would not touch TSM with a 10 foot pole.
The risk exposures of a portfolio are not the sum or union of the risk exposures of components of the portfolio, and this certainly applies to equity factors and the market portfolio.
Last edited by Northern Flicker on Wed Jan 26, 2022 3:10 pm, edited 2 times in total.
-
- Posts: 15369
- Joined: Fri Apr 10, 2015 12:29 am
Re: Fama and French: The Five-Factor Model Revisited
It is well established for small growth. The vast majority of small growth stocks have low returns and a very small minority do exceptionally well. We don't know which of those stocks will do well so as an asset class tilting towards small growth doesn't make sense. Over the long term most small value stocks beat the market so tilting toward them makes sense.Northern Flicker wrote: ↑Wed Jan 26, 2022 12:48 pmIt has not been established that investors overpay for growth. It is one theory.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
Re: Fama and French: The Five-Factor Model Revisited
No, over the long term small value as an asset class outperforms TSM, and just as importantly is less correlated with the total market than small growth. So you don't get any performance benefit by slanting towards small growth, but slanting towards small value improves the consistency of returns in addition to improving overall return.
You've been shown this many times. At this point I don't know if you really don't understand or if you're just trolling.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
Re: Fama and French: The Five-Factor Model Revisited
You are responding to the wrong comment. It is well known that any time there is outperformance by small value, it is due to only a small fraction of the small value stocks outperforming spectacularly. In fact, I read recently that that might be true for the whole stock market - most of the gains are due to only a small fraction of the stocks, which would be why it is better to buy the haystack than search for the few needles.Apathizer wrote: ↑Wed Jan 26, 2022 6:42 pmNo, over the long term small value as an asset class outperforms TSM, and just as importantly is less correlated with the total market than small growth. So you don't get any performance benefit by slanting towards small growth, but slanting towards small value improves the consistency of returns in addition to improving overall return.
You've been shown this many times. At this point I don't know if you really don't understand or if you're just trolling.
Re: Fama and French: The Five-Factor Model Revisited
We do know that SV outperforms TSM over time and SV is less correlated than SG.rkhusky wrote: ↑Wed Jan 26, 2022 8:55 pmYou are responding to the wrong comment. It is well known that any time there is outperformance by small value, it is due to only a small fraction of the small value stocks outperforming spectacularly. In fact, I read recently that that might be true for the whole stock market - most of the gains are due to only a small fraction of the stocks, which would be why it is better to buy the haystack than search for the few needles.Apathizer wrote: ↑Wed Jan 26, 2022 6:42 pmNo, over the long term small value as an asset class outperforms TSM, and just as importantly is less correlated with the total market than small growth. So you don't get any performance benefit by slanting towards small growth, but slanting towards small value improves the consistency of returns in addition to improving overall return.
You've been shown this many times. At this point I don't know if you really don't understand or if you're just trolling.
https://www.portfoliovisualizer.com/bac ... ion2_2=100
We also know expensive, low-profit stocks have lower returns that the overall market, so it makes sense to remove these. That's all I'm arguing. A portfolio with a light-moderate slant towards stocks that are likely to have higher returns. Of course investing in the global market is simpler and will probably meet your goals just fine too. And honestly, I'm beginning to think psychologically that's a better way to invest. Simple, diverse automatically re balancing funds like Life Strategy or target date are probably the best option for most investors. After awhile, trying to figure the 'perfect' allocation can be so obsessive it's annoying.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
-
- Posts: 15369
- Joined: Fri Apr 10, 2015 12:29 am
Re: Fama and French: The Five-Factor Model Revisited
That is different from overpaying, which refers to taking risk not compensated by expected return.Apathizer wrote: ↑Wed Jan 26, 2022 4:43 pmIt is well established for small growth. The vast majority of small growth stocks have low returns and a very small minority do exceptionally well.Northern Flicker wrote: ↑Wed Jan 26, 2022 12:48 pmIt has not been established that investors overpay for growth. It is one theory.
Re: Fama and French: The Five-Factor Model Revisited
We know that it has outperformed in some periods in the past, but will it in the future?Apathizer wrote: ↑Wed Jan 26, 2022 10:08 pm We do know that SV outperforms TSM over time and SV is less correlated than SG.
https://www.portfoliovisualizer.com/bac ... ion2_2=100
Here is a comparable chart, but over the existence of DFA SCV:
https://www.portfoliovisualizer.com/bac ... ion2_2=100
Looks to me like periods of outperformance and underperformance compared to VTSMX. The two came to a tie at the end of March 2020, stayed that way until October 2020, and then SCV had a good year in 2021. Perhaps SCV will outperform for a few years, but then is likely to give it all back at some point. Not really a compelling argument for future outperformance.
And the correlation is time dependent too. Over the last 20 years SCV has had about the same correlation with TSM as SCG. But correlation by itself is pretty useless - what you want to see is how it performs in your portfolio. And the combination of SCG with TSM has yielded better results than SCV + TSM over the last 20 years. But again, who knows what the future holds?
Re: Fama and French: The Five-Factor Model Revisited
To cover all bases I suppose one could hold the TSM and a small cap index like VB, but all that seems to do is slightly increase volatility without increasing expected return. But I guess you never know.rkhusky wrote: ↑Thu Jan 27, 2022 2:43 pmWe know that it has outperformed in some periods in the past, but will it in the future?Apathizer wrote: ↑Wed Jan 26, 2022 10:08 pm We do know that SV outperforms TSM over time and SV is less correlated than SG.
https://www.portfoliovisualizer.com/bac ... ion2_2=100
Here is a comparable chart, but over the existence of DFA SCV:
https://www.portfoliovisualizer.com/bac ... ion2_2=100
Looks to me like periods of outperformance and underperformance compared to VTSMX. The two came to a tie at the end of March 2020, stayed that way until October 2020, and then SCV had a good year in 2021. Perhaps SCV will outperform for a few years, but then is likely to give it all back at some point. Not really a compelling argument for future outperformance.
And the correlation is time dependent too. Over the last 20 years SCV has had about the same correlation with TSM as SCG. But correlation by itself is pretty useless - what you want to see is how it performs in your portfolio. And the combination of SCG with TSM has yielded better results than SCV + TSM over the last 20 years. But again, who knows what the future holds?
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
Re: Fama and French: The Five-Factor Model Revisited
That’s probably what most do - not many willing to go all in on SCV. And maybe you could catch a rebalancing bonus to boot.
Re: Fama and French: The Five-Factor Model Revisited
Small caps in their entirety don't appear to improve performance or lower risk. The track the overall market very closely, so without factor screening they don't seem to provide any practical benefit.
https://www.portfoliovisualizer.com/bac ... ation3_2=8
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
Re: Fama and French: The Five-Factor Model Revisited
+1000 , honestly my view is these systems … which of course always have their creators hawking some funds they manage, is just the latest fade in a long line of them which have never held up. And to which their creators benefit from management fees, and make excuses as to why they didn’t pan out.nisiprius wrote: ↑Sat Jan 15, 2022 11:44 am I have a huge problem with factor advocates, or anyone, saying a) that you have to be prepared to commit for decades to some strategy, while b) changing the strategy a couple of times a decade.
First, "Quality" is not one of the five Fama-French factors, which are "market," "size," "value," "profitability," and "investment."
Second, the constant changes in the story tell us is that financial researchers do not know when they have enough data to make actionable predictions.
This is accompanied by rhetorical maneuvers that walk back claims in subtle ways. And, of course, to be fair, it is not exactly the same people making the initial claims as the people saying later "we never said that."
As Robin Wigglesworth's book, Trillions, confirms, Dimensional Fund Advisors was founded on the size factor. The pure, unadorned size factor and nothing else, and a fund, DFSCX, now known as the DFA US Micro Cap Fund, to allow investors to reap the supposed benefits.
The size factor has been walked back progressively--it's important, just not quite as big as Banz thought; it's important, just not as strong as some others; it's important, but only in raw return, not risk-adjusted return; it's not important even in raw return, but we can "resurrect" it as a flavor enhancer for other factors.
And yet people still constantly post asking how to add small-caps (not small-cap value, just small-caps) to an S&P 500 fund.
Forty years after launch, despite a great start for the first three years,* DFSCX hasn't made a penny more than an S&P 500 index fund,
*snip*
I think the financial industry’s goal is to basically make a product which has mechanics/philosophies which appeal to specific cohorts of investors as to extract fees from them. For techies it is robo advisors (for which software is almost a religion, so why wouldn’t you trust it with your retirement?), for those mathematically/statistically minded they create factor investing, for people who fancy themselves as “elite”— heres some ego stroking private equity funds.
Hard pass.
-
- Posts: 6238
- Joined: Mon Apr 11, 2011 12:28 am
Re: Fama and French: The Five-Factor Model Revisited
The best Factor funds have low fees just like an index funds. They are devised to capture the additional premiums and are based on sound academic finance.mrspock wrote: ↑Thu Jan 27, 2022 10:29 pm+1000 , honestly my view is these systems … which of course always have their creators hawking some funds they manage, is just the latest fade in a long line of them which have never held up. And to which their creators benefit from management fees, and make excuses as to why they didn’t pan out.nisiprius wrote: ↑Sat Jan 15, 2022 11:44 am I have a huge problem with factor advocates, or anyone, saying a) that you have to be prepared to commit for decades to some strategy, while b) changing the strategy a couple of times a decade.
First, "Quality" is not one of the five Fama-French factors, which are "market," "size," "value," "profitability," and "investment."
Second, the constant changes in the story tell us is that financial researchers do not know when they have enough data to make actionable predictions.
This is accompanied by rhetorical maneuvers that walk back claims in subtle ways. And, of course, to be fair, it is not exactly the same people making the initial claims as the people saying later "we never said that."
As Robin Wigglesworth's book, Trillions, confirms, Dimensional Fund Advisors was founded on the size factor. The pure, unadorned size factor and nothing else, and a fund, DFSCX, now known as the DFA US Micro Cap Fund, to allow investors to reap the supposed benefits.
The size factor has been walked back progressively--it's important, just not quite as big as Banz thought; it's important, just not as strong as some others; it's important, but only in raw return, not risk-adjusted return; it's not important even in raw return, but we can "resurrect" it as a flavor enhancer for other factors.
And yet people still constantly post asking how to add small-caps (not small-cap value, just small-caps) to an S&P 500 fund.
Forty years after launch, despite a great start for the first three years,* DFSCX hasn't made a penny more than an S&P 500 index fund,
*snip*
I think the financial industry’s goal is to basically make a product which has mechanics/philosophies which appeal to specific cohorts of investors as to extract fees from them. For techies it is robo advisors (for which software is almost a religion, so why wouldn’t you trust it with your retirement?), for those mathematically/statistically minded they create factor investing, for people who fancy themselves as “elite”— heres some ego stroking private equity funds.
Hard pass.
They are not devised to extract exorbitant fees from investors like PE or HF.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
Re: Fama and French: The Five-Factor Model Revisited
At least it was that way for the 2 year period of your chart.Apathizer wrote: ↑Thu Jan 27, 2022 10:13 pm Small caps in their entirety don't appear to improve performance or lower risk. The track the overall market very closely, so without factor screening they don't seem to provide any practical benefit.
https://www.portfoliovisualizer.com/bac ... ation3_2=8
This chart shows a different story for TSM, SCG, and SCV (using rebalancing bands):
https://www.portfoliovisualizer.com/bac ... tion3_3=50
If you look at the Rolling Returns tab, you can see that each outperforms at different times.
Re: Fama and French: The Five-Factor Model Revisited
Interesting. SC growth and value provided similar benefits, so a SC blend would've provided a similar benefit as well.rkhusky wrote: ↑Fri Jan 28, 2022 7:07 amAt least it was that way for the 2 year period of your chart.Apathizer wrote: ↑Thu Jan 27, 2022 10:13 pm Small caps in their entirety don't appear to improve performance or lower risk. The track the overall market very closely, so without factor screening they don't seem to provide any practical benefit.
https://www.portfoliovisualizer.com/bac ... ation3_2=8
This chart shows a different story for TSM, SCG, and SCV (using rebalancing bands):
https://www.portfoliovisualizer.com/bac ... tion3_3=50
If you look at the Rolling Returns tab, you can see that each outperforms at different times.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
Re: Fama and French: The Five-Factor Model Revisited
Small-Cap Growth can be very good if the lottery stocks are filtered out. My favorite fund for years was a Mid-Cap Growth fund at American Century, I also liked and investing in their Small Cap Growth fund which also has done well. It is in the category of "Tiger in the Tank" investments to add to your portfolio to add a little excitement and hopefully performance as well. REITs, Emerging Markets, International Small-Cap, and Small Value are also aggressive investments worth considering for investment.Apathizer wrote: ↑Fri Jan 28, 2022 3:52 pmInteresting. SC growth and value provided similar benefits, so a SC blend would've provided a similar benefit as well.rkhusky wrote: ↑Fri Jan 28, 2022 7:07 amAt least it was that way for the 2 year period of your chart.Apathizer wrote: ↑Thu Jan 27, 2022 10:13 pm Small caps in their entirety don't appear to improve performance or lower risk. The track the overall market very closely, so without factor screening they don't seem to provide any practical benefit.
https://www.portfoliovisualizer.com/bac ... ation3_2=8
This chart shows a different story for TSM, SCG, and SCV (using rebalancing bands):
https://www.portfoliovisualizer.com/bac ... tion3_3=50
If you look at the Rolling Returns tab, you can see that each outperforms at different times.
A fool and his money are good for business.
Re: Fama and French: The Five-Factor Model Revisited
I'm pretty happy with my current portfolio. While it's only been about two years, the Avantis all-cap funds I own have modestly out-performed their benchmark. Considering value/small value significantly under-performed in 2020 and out-performed in 2021 I'm cautiously optimistic I can expect to consistently beat the market by about 1-2% annually. Only time will tell...nedsaid wrote: ↑Fri Jan 28, 2022 8:33 pmSmall-Cap Growth can be very good if the lottery stocks are filtered out. My favorite fund for years was a Mid-Cap Growth fund at American Century, I also liked and investing in their Small Cap Growth fund which also has done well. It is in the category of "Tiger in the Tank" investments to add to your portfolio to add a little excitement and hopefully performance as well. REITs, Emerging Markets, International Small-Cap, and Small Value are also aggressive investments worth considering for investment.Apathizer wrote: ↑Fri Jan 28, 2022 3:52 pmInteresting. SC growth and value provided similar benefits, so a SC blend would've provided a similar benefit as well.rkhusky wrote: ↑Fri Jan 28, 2022 7:07 amAt least it was that way for the 2 year period of your chart.Apathizer wrote: ↑Thu Jan 27, 2022 10:13 pm Small caps in their entirety don't appear to improve performance or lower risk. The track the overall market very closely, so without factor screening they don't seem to provide any practical benefit.
https://www.portfoliovisualizer.com/bac ... ation3_2=8
This chart shows a different story for TSM, SCG, and SCV (using rebalancing bands):
https://www.portfoliovisualizer.com/bac ... tion3_3=50
If you look at the Rolling Returns tab, you can see that each outperforms at different times.
https://www.portfoliovisualizer.com/bac ... tion4_2=15
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
Re: Fama and French: The Five-Factor Model Revisited
I will say that Avantis has been hitting it out of the park.Apathizer wrote: ↑Fri Jan 28, 2022 9:50 pmI'm pretty happy with my current portfolio. While it's only been about two years, the Avantis all-cap funds I own have modestly out-performed their benchmark. Considering value/small value significantly under-performed in 2020 and out-performed in 2021 I'm cautiously optimistic I can expect to consistently beat the market by about 1-2% annually. Only time will tell...nedsaid wrote: ↑Fri Jan 28, 2022 8:33 pmSmall-Cap Growth can be very good if the lottery stocks are filtered out. My favorite fund for years was a Mid-Cap Growth fund at American Century, I also liked and investing in their Small Cap Growth fund which also has done well. It is in the category of "Tiger in the Tank" investments to add to your portfolio to add a little excitement and hopefully performance as well. REITs, Emerging Markets, International Small-Cap, and Small Value are also aggressive investments worth considering for investment.Apathizer wrote: ↑Fri Jan 28, 2022 3:52 pmInteresting. SC growth and value provided similar benefits, so a SC blend would've provided a similar benefit as well.rkhusky wrote: ↑Fri Jan 28, 2022 7:07 amAt least it was that way for the 2 year period of your chart.Apathizer wrote: ↑Thu Jan 27, 2022 10:13 pm Small caps in their entirety don't appear to improve performance or lower risk. The track the overall market very closely, so without factor screening they don't seem to provide any practical benefit.
https://www.portfoliovisualizer.com/bac ... ation3_2=8
This chart shows a different story for TSM, SCG, and SCV (using rebalancing bands):
https://www.portfoliovisualizer.com/bac ... tion3_3=50
If you look at the Rolling Returns tab, you can see that each outperforms at different times.
https://www.portfoliovisualizer.com/bac ... tion4_2=15
The weird thing is that I have had investments with American Century for years but haven't yet tried their Avantis ETFs yet. I am getting interested.
A fool and his money are good for business.
Re: Fama and French: The Five-Factor Model Revisited
Thanks for the links. It seems that the analysis is for each factor independently?Apathizer wrote: ↑Tue Jan 18, 2022 12:26 pmHere are links to a detailed paper and a video summarizing the research.wenchleaf wrote: ↑Tue Jan 18, 2022 11:57 amAny more info (sources or discussion) on this?Apathizer wrote: ↑Sat Jan 15, 2022 11:55 amFremdon Ferndock wrote: ↑Sat Jan 15, 2022 9:19 amhint: it isn't value and it isn't smallSo how well has Fama and French’s five-factor model explained returns over the decades? According to our analysis, only one factor has truly held up over all time periods.
https://blogs.cfainstitute.org/investor ... revisited/Size is especially weak in ex-US and emerging markets.drumboy256 wrote: ↑Sat Jan 15, 2022 10:37 am This goes to show that Avantis is ahead of the curve again in terms of quality. Not surprising really.
https://www.pwlcapital.com/resources/fi ... with-etfs/
https://www.youtube.com/watch?v=jKWbW7Wgm0w
In the paper and on the Rational Reminder podcast, it's stated that size is weak on its own but has "powerful interaction effects between size and other factors, such as value" and cites Blitz and Hanauer 2020.
I haven't looked at the Blitz, Hanauer paper yet, but it does appear, that DGS does outperform DFEVX (essentially AVES) over the period of available data on portfolio visualizer.
DGS is Emerging Market Small Cap Dividend, so some loading on Value in the regression, and DFEVX/AVES of course are Emerging Market Value. DFEVX does have a bit higher expense ratio than AVES, but the CAGR differences are beyond that. And of course, PV does incorporate DGS's higher expense over DFEVX. You can also add VWO in there for reference, which ends up between the two.
-
- Posts: 605
- Joined: Wed May 28, 2014 11:53 am
Re: Fama and French: The Five-Factor Model Revisited
DGS doesn't really load on anything other than SmL. No meaningful value loading, which is not surprising given the somewhat weak link between dividend and value. DEMSX is DFA's small cap emerging markets fund. The performance of DGS and DEMSX is very similar.wenchleaf wrote: ↑Sat Jan 29, 2022 2:49 am DGS is Emerging Market Small Cap Dividend, so some loading on Value in the regression, and DFEVX/AVES of course are Emerging Market Value. DFEVX does have a bit higher expense ratio than AVES, but the CAGR differences are beyond that. And of course, PV does incorporate DGS's higher expense over DFEVX. You can also add VWO in there for reference, which ends up between the two.
Re: Fama and French: The Five-Factor Model Revisited
As I understand it, and that's a major caveat since I'm not an expert by any means, factors do tend to be stronger in small caps, but the effect is still fairly small and applies mainly to SV. For instance, in emerging markets SMB is 0.71 and HML is 6.67. If I understand correctly, this means you're only getting an additional 0.71% of expect return for SV relative to HML for the entire market. The size effect is a little larger in US and developed markets, but it's still the smallest factor.wenchleaf wrote: ↑Sat Jan 29, 2022 2:49 amThanks for the links. It seems that the analysis is for each factor independently?Apathizer wrote: ↑Tue Jan 18, 2022 12:26 pmHere are links to a detailed paper and a video summarizing the research.wenchleaf wrote: ↑Tue Jan 18, 2022 11:57 amAny more info (sources or discussion) on this?Apathizer wrote: ↑Sat Jan 15, 2022 11:55 amFremdon Ferndock wrote: ↑Sat Jan 15, 2022 9:19 am
hint: it isn't value and it isn't small
https://blogs.cfainstitute.org/investor ... revisited/Size is especially weak in ex-US and emerging markets.drumboy256 wrote: ↑Sat Jan 15, 2022 10:37 am This goes to show that Avantis is ahead of the curve again in terms of quality. Not surprising really.
https://www.pwlcapital.com/resources/fi ... with-etfs/
https://www.youtube.com/watch?v=jKWbW7Wgm0w
In the paper and on the Rational Reminder podcast, it's stated that size is weak on its own but has "powerful interaction effects between size and other factors, such as value" and cites Blitz and Hanauer 2020.
I haven't looked at the Blitz, Hanauer paper yet, but it does appear, that DGS does outperform DFEVX (essentially AVES) over the period of available data on portfolio visualizer.
DGS is Emerging Market Small Cap Dividend, so some loading on Value in the regression, and DFEVX/AVES of course are Emerging Market Value. DFEVX does have a bit higher expense ratio than AVES, but the CAGR differences are beyond that. And of course, PV does incorporate DGS's higher expense over DFEVX. You can also add VWO in there for reference, which ends up between the two.
It's also important to consider the 5 factors seem uncorrelated with each other, so they're likely to emerge unpredictably at random times. That's why it seems important to be broadly diversified across all 5 factors. SMB might be absent for awhile while HML surges. Or vice versa. The same is true for any of the 5.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
Re: Fama and French: The Five-Factor Model Revisited
Your comments on small cap being uncorrelated only go directly against the PWL analysis paper you linked, specifically the statement I put in quotes, and Ben Felix's comments on his podcast as I mentioned. What evidence do you have for that?Apathizer wrote: ↑Sat Jan 29, 2022 1:03 pmAs I understand it, and that's a major caveat since I'm not an expert by any means, factors do tend to be stronger in small caps, but the effect is still fairly small and applies mainly to SV. For instance, in emerging markets SMB is 0.71 and HML is 6.67. If I understand correctly, this means you're only getting an additional 0.71% of expect return for SV relative to HML for the entire market. The size effect is a little larger in US and developed markets, but it's still the smallest factor.wenchleaf wrote: ↑Sat Jan 29, 2022 2:49 amThanks for the links. It seems that the analysis is for each factor independently?Apathizer wrote: ↑Tue Jan 18, 2022 12:26 pmHere are links to a detailed paper and a video summarizing the research.
https://www.pwlcapital.com/resources/fi ... with-etfs/
https://www.youtube.com/watch?v=jKWbW7Wgm0w
In the paper and on the Rational Reminder podcast, it's stated that size is weak on its own but has "powerful interaction effects between size and other factors, such as value" and cites Blitz and Hanauer 2020.
I haven't looked at the Blitz, Hanauer paper yet, but it does appear, that DGS does outperform DFEVX (essentially AVES) over the period of available data on portfolio visualizer.
DGS is Emerging Market Small Cap Dividend, so some loading on Value in the regression, and DFEVX/AVES of course are Emerging Market Value. DFEVX does have a bit higher expense ratio than AVES, but the CAGR differences are beyond that. And of course, PV does incorporate DGS's higher expense over DFEVX. You can also add VWO in there for reference, which ends up between the two.
It's also important to consider the 5 factors seem uncorrelated with each other, so they're likely to emerge unpredictably at random times. That's why it seems important to be broadly diversified across all 5 factors. SMB might be absent for awhile while HML surges. Or vice versa. The same is true for any of the 5.
From the podcast transcript
Benjamin Felix: Here's an interesting one. So no. Size hasn't really been established as an independent risk factor, which some people may be surprised to hear. There's no real good theoretical basis for a standalone size premium. I'd say it's weak at best. And in terms of statistical reliability, it hasn't been really close. I mean, somewhat close, closer than something random, I guess. But the T stats have not been nearly as high as something like value or the equity risk premium in historical data.
He does talk a bit more about it, including referencing some other papers, in the link/podcast. But yes, from what I understand, the other factors are more independent.
As for the values, as I mentioned, I believe they SMB and HML are all calculated independently, though the paper mentioned in the PWL analysis may consider the interdependencies.
That is essentially part of what I meant to convey, but you have done a better job. It loads up on small cap, obviously given its name, and some of value, given value's weak (but not zero) correlation to dividends. The point was to show that small cap increases the return on value. DEMSX may have been a better fit to show this, though it's not easily accessible compared to AVES (which is essentially DFEVX), it does load up more on small and a bit less on value compared to DGS.DaufuskieNate wrote: ↑Sat Jan 29, 2022 10:26 amDGS doesn't really load on anything other than SmL. No meaningful value loading, which is not surprising given the somewhat weak link between dividend and value. DEMSX is DFA's small cap emerging markets fund. The performance of DGS and DEMSX is very similar.wenchleaf wrote: ↑Sat Jan 29, 2022 2:49 am DGS is Emerging Market Small Cap Dividend, so some loading on Value in the regression, and DFEVX/AVES of course are Emerging Market Value. DFEVX does have a bit higher expense ratio than AVES, but the CAGR differences are beyond that. And of course, PV does incorporate DGS's higher expense over DFEVX. You can also add VWO in there for reference, which ends up between the two.
Re: Fama and French: The Five-Factor Model Revisited
I think I either misunderstood or mis-explained. I thought SMB referred to SV equities, but I guess SMB just refers to size and nothing else? If that's the case, then yes, as Ben says SMB seems weak, especially in ex-US markets.wenchleaf wrote: ↑Sat Jan 29, 2022 2:12 pmYour comments on small cap being uncorrelated only go directly against the PWL analysis paper you linked, specifically the statement I put in quotes, and Ben Felix's comments on his podcast as I mentioned. What evidence do you have for that?
From the podcast transcript
Benjamin Felix: Here's an interesting one. So no. Size hasn't really been established as an independent risk factor, which some people may be surprised to hear. There's no real good theoretical basis for a standalone size premium. I'd say it's weak at best. And in terms of statistical reliability, it hasn't been really close. I mean, somewhat close, closer than something random, I guess. But the T stats have not been nearly as high as something like value or the equity risk premium in historical data.
This also reinforces why he and so many other recommend simple cap-weight total market index funds for most investors. Unless someone is really focused on investing as a hobby it's really easy to fall down an obsessive rabbit hole trying to configure the best portfolio. Most people have other things they want focus on, so a single, simple auto re-balancing fund is probably the best option for most investors.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
Re: Fama and French: The Five-Factor Model Revisited
I believe you misunderstood. SMB indeed just means size, it means Small Minus Big, just like HML means High Minus Low. These models do actually refer to shorting Big Size and Low Value hence Minus. I assume by SV you mean small and value?Apathizer wrote: ↑Sat Jan 29, 2022 3:39 pmI think I either misunderstood or mis-explained. I thought SMB referred to SV equities, but I guess SMB just refers to size and nothing else? If that's the case, then yes, as Ben says SMB seems weak, especially in ex-US markets.wenchleaf wrote: ↑Sat Jan 29, 2022 2:12 pmYour comments on small cap being uncorrelated only go directly against the PWL analysis paper you linked, specifically the statement I put in quotes, and Ben Felix's comments on his podcast as I mentioned. What evidence do you have for that?
From the podcast transcript
Benjamin Felix: Here's an interesting one. So no. Size hasn't really been established as an independent risk factor, which some people may be surprised to hear. There's no real good theoretical basis for a standalone size premium. I'd say it's weak at best. And in terms of statistical reliability, it hasn't been really close. I mean, somewhat close, closer than something random, I guess. But the T stats have not been nearly as high as something like value or the equity risk premium in historical data.
With that misunderstanding explained, do the quotes I linked from Ben Felix, PWL, and the PV link make more sense then?
The Blitz, Hanauer 2020 paper may look at some of the interactions between Size and other factors, and I'm not sure if other papers have as well.
Agreed, I would never tell any friends to do factor investing. I always have found academic papers and research interesting in many fields, and recently this has extended to finance, for better or worse.Apathizer wrote: ↑Sat Jan 29, 2022 3:39 pmThis also reinforces why he and so many other recommend simple cap-weight total market index funds for most investors. Unless someone is really focused on investing as a hobby it's really easy to fall down an obsessive rabbit hole trying to configure the best portfolio. Most people have other things they want focus on, so a single, simple auto re-balancing fund is probably the best option for most investors.
-
- Posts: 15369
- Joined: Fri Apr 10, 2015 12:29 am
Re: Fama and French: The Five-Factor Model Revisited
NAESX tracked the Russell 2000 Index through May 16, 2003; MSCI US Small Cap 1750 Index through January 30, 2013; CRSP US Small Cap Index thereafter.
The CRSP SC index is about 1/3 midcaps and 2/3 smallcaps using any non-CRSP taxonomy.
IJR, IJS, and IJT would be a good comparison:
https://www.portfoliovisualizer.com/bac ... ion3_3=100
The quality screens of IJT avoid the problematic irrational segment of SCG (microcap "lottery tickets").
Last edited by Northern Flicker on Sun Jan 30, 2022 12:22 am, edited 1 time in total.
Re: Fama and French: The Five-Factor Model Revisited
Yes it does, thanks, and yes SV is small-cap value abbreviated. Since size is comparatively the least significant factor, it seems to make sense to focus on the more significant factors like HML, CMA, and RMW. Then again some people much smarter than me think those factors are largely time-dependent, so who knows?wenchleaf wrote: ↑Sat Jan 29, 2022 6:40 pmI believe you misunderstood. SMB indeed just means size, it means Small Minus Big, just like HML means High Minus Low. These models do actually refer to shorting Big Size and Low Value hence Minus. I assume by SV you mean small and value?
With that misunderstanding explained, do the quotes I linked from Ben Felix, PWL, and the PV link make more sense then?
I wrote: ↑Sat Jan 29, 2022 3:39 pmThis also reinforces why he and so many others recommend simple cap-weight total market index funds for most investors. Unless someone is really focused on investing as a hobby it's really easy to fall down an obsessive rabbit hole trying to configure the best portfolio. Most people have other things they want focus on, so a single, simple auto re-balancing fund is probably the best option for most investors.
LOL
Me too. I'm highly scientific, but until recently haven't studied investing science much, also for better/worse. As with so many other fields, it's really complicated and sometimes it seems pursuit of perfection might be more trouble than it's worth. As you might know, there's evidence showing those money-focused tend to be less happy. Also, maximizers, that is people more focused on trying to find the best option, are less happy than satisficers, that is people who are content with good enough.
Considering this, factor investing might be self defeating. Sure, it might lead to about 1-3% higher returns, but I don't know if the complexity is worth it. I'm happy with my Avantis portfolio so far, but really I should probably move everything into a Life Strategy or Target Date Fund since they'd meet my goals just fine with less fretting and complexity.
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.
-
- Posts: 5561
- Joined: Fri Feb 23, 2007 7:21 pm
Re: Fama and French: The Five-Factor Model Revisited
Just couple comments.Apathizer wrote: ↑Sat Jan 29, 2022 9:52 pmYes it does, thanks, and yes SV is small-cap value abbreviated. Since size is comparatively the least significant factor, it seems to make sense to focus on the more significant factors like HML, CMA, and RMW. Then again some people much smarter than me think those factors are largely time-dependent, so who knows?wenchleaf wrote: ↑Sat Jan 29, 2022 6:40 pmI believe you misunderstood. SMB indeed just means size, it means Small Minus Big, just like HML means High Minus Low. These models do actually refer to shorting Big Size and Low Value hence Minus. I assume by SV you mean small and value?
With that misunderstanding explained, do the quotes I linked from Ben Felix, PWL, and the PV link make more sense then?
I wrote: ↑Sat Jan 29, 2022 3:39 pmThis also reinforces why he and so many others recommend simple cap-weight total market index funds for most investors. Unless someone is really focused on investing as a hobby it's really easy to fall down an obsessive rabbit hole trying to configure the best portfolio. Most people have other things they want focus on, so a single, simple auto re-balancing fund is probably the best option for most investors.LOL
Me too. I'm highly scientific, but until recently haven't studied investing science much, also for better/worse. As with so many other fields, it's really complicated and sometimes it seems pursuit of perfection might be more trouble than it's worth. As you might know, there's evidence showing those money-focused tend to be less happy. Also, maximizers, that is people more focused on trying to find the best option, are less happy than satisficers, that is people who are content with good enough.
Considering this, factor investing might be self defeating. Sure, it might lead to about 1-3% higher returns, but I don't know if the complexity is worth it. I'm happy with my Avantis portfolio so far, but really I should probably move everything into a Life Strategy or Target Date Fund since they'd meet my goals just fine with less fretting and complexity.
1. The other factors may have bigger effect in small stocks, so a big reason to focus on small is to see more bang for the buck from other factors.
2. In the set it and forget it / satisfice realm, I’d be curious about your thoughts after seeing the charts in Chapter 9 of Larry Swedroe’s Factor book. They basically show that a portfolio diversified across factors has been less likely to have a negative outcome over any time period 1-20 years. I could potentially see a satisficer being happier with a portfolio tilted to the factors on the equity side and having overall lower equity allocation. His portfolio will always have components over performing or under performing a more TSM portfolio, but the ride could be smoother and perhaps increased complexity results in less fretting. Bet you know all this already, interested in your thoughts.
Dave
Re: Fama and French: The Five-Factor Model Revisited
1. True, but it does appear fairly small. A relatively moderate overweighting in SV probably makes sense, but going way overboard doesn't since it greatly increases risk.Random Walker wrote: ↑Sun Jan 30, 2022 1:03 pm Just couple comments.
1. The other factors may have bigger effect in small stocks, so a big reason to focus on small is to see more bang for the buck from other factors.
2. In the set it and forget it / satisfice realm, I’d be curious about your thoughts after seeing the charts in Chapter 9 of Larry Swedroe’s Factor book. They basically show that a portfolio diversified across factors has been less likely to have a negative outcome over any time period 1-20 years. I could potentially see a satisficer being happier with a portfolio tilted to the factors on the equity side and having overall lower equity allocation. His portfolio will always have components over performing or under performing a more TSM portfolio, but the ride could be smoother and perhaps increased complexity results in less fretting. Bet you know all this already, interested in your thoughts.
Dave
2. Yes, as I've said I'd be happy if there were a widely available globally diversified single fund option with broad, moderate factor slants like DFA Global Equity. I don't know of any options right now though. Also, as one gets nearer retirement and allocates more to bonds factor slants produce diminishing returns. If you compare DFA 60/40 global allocation with Vanguard Life Strategy Moderate Growth (also a 60/40 allocation), there's no statistically significant performance difference. At that point factor fretting seems especially pointless, so factors are more relevant in the long-term.
https://www.portfoliovisualizer.com/bac ... ion2_2=100
https://www.portfoliovisualizer.com/bac ... ion2_2=100
ROTH: 50% AVGE, 10% DFAX, 40% BNDW. Taxable: 50% BNDW, 40% AVGE, 10% DFAX.