Seeking Alpha

SDS (Seductive ...'s  Instablog

SDS (Seductive Dividend Stocks)
Send Message
Sorry I hide my true identity but I'm a physicist/engineer, native contrarian and idea generator. I am an eclectic dividend investor with motto "In God We Trust, All Others Pay Cash" applied to companies I invest in. I like to read /and read a lot - did you look on my SA photo 8-)? /... More
My blog:
SDS blog in Seeking Alpha
  • A Funny Note On MPT Inspiered By Larry Swedroe Vs. Amalogoli Discussion (Feb. 1, 2012) 48 comments
    Feb 2, 2012 2:24 PM
    This Note inspiered by comments to the good SA article "Larry Swedroe (Passively) Positions For 2012: Avoiding Stock And Sector Picking, Economic Forecasting, And All Other Forms Of Speculation"

    Modern portfolio theory (MPT) is cornestone of traditional finance. I'm a simple minded person. To me MPT is an attempt to show that "investing is a tradeoff between risk and expected return" (bit.ly/yxHZ1z). According to the same source (bit.ly/xeb0c0) "In finance, risk is the probability that an investment's actual return will be different than expected." If I combine these 2 definition I get investing is a tradeoff between real and expected returns. Hence, if my real return is equal to my expected return I do not invest.

    Let me give an example. When I buy a stock at 10$ and decide to sell it at 25$ my expected return is 15$. When later on I indeed sell the stock at 25$ my real return is 15$. Fine I happy with 15$ and because I did not invest according to the definitions.
    Aliluya, now I know that to tell IRS about my capital gains !!!!

    Well return can be measured (15$) and risk is not. So the MPT attempt is not successful (at least for me as a physicist). OK, at least do not call it " theory ".

    BTW, the discussion and article (see seekingalpha.com/article/315875-larry-sw...) are quite good.

Back To SDS (Seductive Dividend Stocks)'s Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

Comments (48)
Track new comments
  • varan
    , contributor
    Comments (3704) | Send Message
     
    In the definition of risk that you are talking about, 'expected' does not mean what you think it means.
    2 Feb 2012, 02:27 AM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Well I took wikipedea definitions.

     

    English is not my native so I checked on Google (http://bit.ly/zA2ZHA) and got

     

    "expected past participle, past tense of expect (Verb)Verb: Regard (something) as likely to happen.
    Regard (someone) as likely to do or be something."

     

    What is "expected" in your mind?
    2 Feb 2012, 02:29 PM Reply Like
  • varan
    , contributor
    Comments (3704) | Send Message
     
    Pick up any book on probability and statistics.
    2 Feb 2012, 08:03 PM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Look my bio @ SA. I know stat & use JMP.
    3 Feb 2012, 01:23 AM Reply Like
  • varan
    , contributor
    Comments (3704) | Send Message
     
    then you should know what 'expected' means.
    3 Feb 2012, 01:57 AM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » I asked you to define. Expected in engineering might be different from expected in finance. But in both (and all other) fields a theory must explain the simplest case. Let look again:

     

    "In finance, risk is the probability that an investment's actual return will be different than expected." In simplest case: risk = (actual return - expected return).

     

    "investing is a tradeoff between risk and expected return". In simplest case: investing = risk/reward = (actual return - expected return)/reward.

     

    When actual return = expected return, risk is zero and I did not invest just was rewarded.

     

    Any lottery big winner did not face any risk because expected reward was announced before lottery!?

     

    Can you disproof?
    3 Feb 2012, 02:22 AM Reply Like
  • varan
    , contributor
    Comments (3704) | Send Message
     
    For a lottery:

     

    Expected Reward = Probability of Winning the lottery*Lottery Prize - Probability of not winning the lottery*the price of lottery ticket.
    3 Feb 2012, 09:56 AM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Because I can claim "expected return" after I got real return (e.g. I expected to win 1M$ in lottery and I won)
    "Aliluya, now I know that to tell IRS about my capital gains !!!!"
    Disclaimer: I do not play any lottery
    3 Feb 2012, 10:38 AM Reply Like
  • varan
    , contributor
    Comments (3704) | Send Message
     
    OK let us go back to MPT.

     

    You can start here.

     

    http://bit.ly/zu9Byf

     

    You are confused between 'possible' return and 'expected' return.

     

    Since you are a physicist, here is the definition of expected return:

     

    Integral[r*f(r),-inf,+...

     

    where r is the return, and f(r) its probability density function. This is what the theory means when it talks about 'expected' return. You estimate f(r) from the past data.

     

    Expected return has meaning only before you realize the return, not after.
    3 Feb 2012, 12:28 PM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » OK, thanks for the expected return definition. As far as I know inf is infinite time. I guess a linearly causal arrow of time is presumed in calculations with "inf" in finance. This might be not only the case for infinite time. Also from symmetry & thermodynamics points of view data for infinite past time might be required and we know that such data cannot exist because of 2 simple limits: our Universe has finite history, humans and their financial activities have even shorter history.

     

    I use all definitions literally (English is not my native, so to me expected is the past tense). I provided a simple sample there statistics is not needed. I intentionally avoid probability because as soon as you use it you have mismatch in units for measurements (if reward or return is in $, risk also must be in $ while probability is usually % or unitless number).

     

    I think 'possible' return vs. 'expected' return is just even worse semantics. To me investing = risk/reward = (actual return - expected return)/reward sounds better than investing = risk/reward = (actual return - possible return)/reward because set of a rational investor's expectations is smaller than set of all possibilities. While the set of expectations does not include Black Swans by definition, these "birds" are in set of all possibilities (see "The Black Swan" by Nassim Nicholas Taleb ).

     

    Actually my initial intention was to make a joke "...now I know that to tell IRS about my capital gains !!!!" but as truth is often behind any good fairy tale as a serious point is often behind any good joke.

     

    A serious point that more knowledgeable than I people have valid concerns about MPT. I as a physicist at least would vote to replace the word "theory" in the title to a word "model", "simulator", "toolbox" , etc... and not put the label "science" just to attract public.

     

    Anyway it is fine if MPT works for Larry Swedroe and other folks. I can live without it....

     

    Amen!
    4 Feb 2012, 02:02 AM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    SDS
    As pointed out your problem is your inorrect definition of expected return--which is not GUARANTEED. It is only the mean of the potential disperion of returns, with many other alternative outcomes possible. If that was not true there would be no risk.
    And of course this relationship is only one part of MPT.
    But I would add that this relationship is part of ALL ECONOMIC THEORY, not just MPT, or better I would suggest using modern financial theory.

     

    Best
    Larry
    13 Feb 2012, 07:36 AM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    SDS
    Note the word theory is appropriate, it is a hypothesis, not a LAW, like gravity. We don't say the theory of gravity.

     

    MPT is a body of work, not just the risk and expected return part. It's also about how asset classes mix with each, not thinking about them in isolation. And it's also about WHEN risks tend to show up. Assets that perform badly in bad times should carry higher risk premiums.

     

    The problem with many of the critics is that they don't know what they don't know, like the example SDS provided on expected return. It was his incorrect use of the term that leads to his critique.

     

    Finally, MPT gives us the best model we have for thinking about how markets work and what strategy is the one most likely to help us achieve our goals. It is also an evolving model, moving from CAPM to three-factor model, to four-factor model and now more align the ideas of multiple sources of return. For those interested suggest reading Antti Illmanen's wonderful Expected Returns.

     

    Best wishes
    Larry
    16 Feb 2012, 09:40 AM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Larry,
    I'm biased by physics terminology. In physics and mathematics theory is well-defined term and when I read a regular MPT paper my instincts reject to consider it as "a theory" similar to let's say quantum mechanics. Just personal..... As soon as you say "It is a model" I kind of accept it better but being a scientists my instincts force me to propose a better model. For example FF 3-factor is obviously destruction of model beauty and better model might make more sense. Again in physics we operate with models but know their limits and sometime unproven assumption we made in order to describe something. A good theory should be bullet proof and not only explain but predict. I don't like to "open kimono" to much but one of my professional achievements was demonstration that one field I work in with long-lasting label "theory" is just methodology. My book about this field is admitted as the best book. Therefore I'm quite sensitive to word "theory".

     

    Because in finance many terms are ill-defined I can play a game with words (similar to many finance pundits pseudo-scientific claims) even in my non-native English. I didn't intend to hurt you or Amalogoli. I just like to make jokes....

     

    SDS
    16 Feb 2012, 11:56 AM Reply Like
  • varan
    , contributor
    Comments (3704) | Send Message
     
    Larry, you should give up. Some are unteachable.
    16 Feb 2012, 12:05 PM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Varan,
    No I'd like to learn and going to read Larry's book.
    I hope you recognize the word FUNNY in the title.
    SDS
    17 Feb 2012, 12:57 AM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    SDS
    Will write this up but high growth in div strategy, top quintile reformed every year, 82-11 beat S&P 500 by just 0.1 percent. Think we can agree this is statistically insignificant difference. Also note the loading on size and value was IDENTICAL, not even virtually, but IDENTICAL to the S&P 500. There is no there there as they say. Want higher returns simply load up on SV, then can take less beta risk if you prefer. Also would be far more diversified by number of stocks and also can do it globally as well
    Larry
    18 Feb 2012, 11:12 AM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Larry,

     

    Yes, I would ignore even at 1% difference. I work too much with quantum fluctuations to know how to ignore such noise.

     

    I think the main point you (even knowing behavior finance) and other critics of dividend investment are missing even if we speak in beta risk terminology (I doubt that it is correct but this is a theme of separate research) is:
    I (and some other DGI) do NOT want higher returns if they are unstable at intermediate ( below ~5 years) time scale. I rather satisfy with modest more consistent positive income (that I can re-invest and retires can take for living) without record spikes in both positive and especially negative directions. (Remember - human reaction to loss is ~ 3X stronger than to gain).
    I use famous in almost any engineering discipline rule of thumb "It takes 20% of effort to achieve 80% of result and it takes 80% of effort to achieve 20% of result". Based on long-term data I initially found in James Montier book "Value Investing: Tools and Techniques for Intelligent Investment" I can get more than 80% of long-term (in my classification above 10 years) returns via dividend components such as yield and growth (see http://seekingalpha.co...) even without value-type screen for high initial yield stocks.

     

    IMO, the cornerstone concept of assets allocation is transfer of previous covariance into future, and again IMO this transfer might be not as straightforward as we like to see it with any financial data.
    The old Wall Street sayings goes as "the only thing that goes up in a bear market is correlation". I'd add from 2008-2011 experience and numbers presented in the tables above another thing that goes up in a bear market is the number of dividend cuts and omissions.
    One more fact against dividend investment I aware of - dividends (in $$$) of several respectful quasi-index mutual funds (I analyzed about dozen from Vanguard, Fidelity, etc) DECREASE slowly but steady with time between ~ 1985 and 2010. In order to protect myself from this trend I invest in DG stocks but not exclusively and tend to buy small-cap dividend stocks at good historical value. So I mimic SV fund with restriction for non-zero dividends and this choice came from prior studies which forced FF wrote 1992 paper (although I'm not sure that these 2 factors will be important in future because now everybody /and I consider myself as a person at the end of informational chain/ knows).

     

    I hope I proof for "varan" with last 2 statements that even I call myself sometimes as a dividend zealot (even I probably use this word historically incorrect - see http://bit.ly/xi4NEX), I am not an idiot who just listen self-named authorities (although I always appreciate ideas and opinions of real experts).

     

    SDS
    18 Feb 2012, 12:19 PM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    SDS
    Basically we are in agreement
    The bottom line is IMO that the dividend growth guys totally ignore the risks that divs can be cut or even eliminated, especially if we have severe economic times. Thus they are NOT in anyway a substitute for high quality FI. This is just IMO are really bad idea.
    Far better to simply tilt the portfolio to SV, and do it globally, and then can lower exposure to beta because of the higher expected return. This cuts the tail risk.
    Also what is missing in the analysis is that not all correlations rise in severe bear markets, the correlation to safe FI goes WAY DOWN, and that helps reduce volatility, and with the higher expected return of SV allowing one to hold more safe FI then you get another benefit there.
    If you look at the data this has been/is far superior strategy.
    The high div growth strategy IMO has nothing to offer that cannot be achieved in more efficient manner.
    Also if the laws change on taxes on divs this will further add to the issues for this strategy

     

    Best wishes
    Larry
    18 Feb 2012, 12:27 PM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Larry,

     

    I collected cuts info and conduct research to minimize my exposure to cuts - see my other instablogs.

     

    Last few months I shopped only in Europe (via ADR, I don't have direct access) and I hold some stocks I call "nano-caps".

     

    "...the correlation to safe FI"... - what is FI?

     

    "If you look at the data this has been/is far superior strategy.
    The high div growth strategy IMO has nothing to offer that cannot be achieved in more efficient manner."
    I'm not aware about a "superior strategy" that allows me to achive the goal I outlined in my previous comment to this post with effort I can handle. Hope to find it in your book.

     

    Agree about taxes - hate them and believe that investors must change the double taxation"rule that exist in USA (Hong Kong prospers very well without double taxation).

     

    Sincerely
    SDS
    18 Feb 2012, 01:10 PM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    fixed income
    The superior strategy is the LARRY portfolio, check the NY TIMES article on it
    or read the appendix in Only Guide to Right Financial Plan, requires that out of box thinking (:-))
    18 Feb 2012, 01:28 PM Reply Like
  • varan
    , contributor
    Comments (3704) | Send Message
     
    Larry
    What would have been the historical return for the LARRY portfolio if someone had implemented it starting, say, in 2000?

     

    Thanks.
    18 Feb 2012, 03:33 PM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Larry
    Please email the article or link.
    SDS
    18 Feb 2012, 04:23 PM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    Varan
    Well the answer depends on how much bonds you hold and whether they were munis or faxables, but for the EQUITY portion it might be something like, (whether stocks in taxable or not (different funds used) and how you rebalanced, but using annual rebalancing
    50% DFA SV
    35% DFA ISV and
    15% EMV
    So that return would have been 10% with SD of 27

     

    SDS have written it but not posted it yet, just become regular reader of my blog
    http://bit.ly/AchOUv;contentBody
    18 Feb 2012, 06:37 PM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » "Also what is missing in the analysis is that not all correlations rise in severe bear markets, the correlation to safe FI=fixed income goes WAY DOWN"
    Larry please read the maxim ""the only thing that goes up in a bear market is correlation" carefully - it is not about correlations, it is about up movement during bear market.
    18 Feb 2012, 08:01 PM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    Exactly
    So the maxim as stated is wrong, what it should say is that the correlation of RISKY assets goes up in bear markets. The correlation of safe to risky assets collapses
    shows the most important diversification is to safe fixed income making the dividend strategies poor diversifiers and certainly not substitutes for fixed income. IT amazes me that people even debate this.

     

    Best wishes
    Larry
    19 Feb 2012, 08:41 AM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Yes, my point was to show that stock prices AND dividends go down in bear market.

     

    50% DFA SV = small value
    35% DFA ISV = international small value (?)
    15% EMV = emerging market value (?)

     

    Well we know that small and value stocks gave high return in the past. What cause them to outperform in the future (please do not cite FF92)?
    19 Feb 2012, 12:02 PM Reply Like
  • varan
    , contributor
    Comments (3704) | Send Message
     
    By the way the main feature of Larry's portfolio that makes its risk adjusted performance measures so attractive is not the small cap portion of it, but the fact that it allocates 70% to essentially risk free FI.

     

    Once you start with that it is not extremely difficult to add other assets (not individual equities but broad based mutual funds or ETFs) - not necessarily small cap value - to the portfolio to match the performance of the Larry portfolio. Inasmuch as the main metrics that Larry uses are the risk adjusted performance measures - three factor alpha, Sharpe ratio, risk adjusted return (Modigilani's M2?) etc. - it appears to me that if you match those measures even without the small cap value it is just as good according to such criteria.

     

    For an individual investor that seems to be one of the only few choices available, since DFA funds appear to be so difficult to buy.
    19 Feb 2012, 12:23 PM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Varan/Larry,

     

    I hold 3years expenses in cash/CD. I'm going to read Larry's ALPHA book to make discussion better.

     

    Meantime let me explain what I'd like to know:
    If I drop a stone from roof it come to ground because of gravity - this is the fundamental source. So my question is:
    Did financial science find a fundamental source why small and value stocks will always outperform grown and large stocks? I'd not buy FF92 arguments - IMO this is attempt to save existing probably incorrect model.

     

    SDS
    19 Feb 2012, 12:55 PM Reply Like
  • varan
    , contributor
    Comments (3704) | Send Message
     
    As a quantum physicist, you should know better than that. The stone will drop, but even in vaccuum, the acceleration will neither be uniform, nor will its history be the same each time you drop it. The Unified Field theory to explain this has not been developed.

     

    The Unified Theory to explain the time history of the stock prices will never be developed.
    19 Feb 2012, 01:26 PM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    SDS
    Very simple the correlation between risk and EXPECTED return, Small and value stocks are stocks of riskier companies and thus MUST have higher expected returns, just as stocks have higher expected returns than bonds, but not guaranteed
    This is the point you seem to keep missing. There is no guarantee. There are plenty of studies as I have stated showing the very simple and logical risk stories.
    Best wishes
    Larry
    19 Feb 2012, 06:24 PM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Larry,
    You are right that expected return isn't guaranteed return. You use FF92 argument with fuzzy terminology about risk for SV premium.
    Everybody knows about SV premium so I don't think it is correct explanation. Recent "abnormal" bonds/stocks yields ratio IMO makes the concept questionable.
    SDS
    20 Feb 2012, 12:29 AM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    I don't know why you think there is funny terminology for risk premium.Nothing fuzzy about it at all. It is a risk premium with risks generated by very common and simple logical explanations, see the studies. And don't know what you think is "abnormal" now. I don't see anything abnormal nor has there been anything abnormal in last several years.
    BTW-the FF model works extremely well at explaining the differences in returns of diversified portfolios, tstats very high, and it works all over world. Is it perfect, NO? The four factor model does even better, adding momentum. Is it a very good model that helps us explain how world works, emphatically yes. Does basically the whole academic community use now the four factor model to analyze markets and investments, yes. It is the standard. So the whole academic community is wrong and you are right? Possible, but ask yourself seriously if that is likely. Note also that FF use the four factor model themselves in analyzing things
    Best wishes
    Larry
    20 Feb 2012, 08:00 AM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Larry,

     

    I consider negative risk premium as abnormal. Absence of stability of bonds yield curve profile is another example that risk is fuzzy rather than robust.

     

    I hope you remember from school that almost any curve can be fitted with 5 parameters model.
    Thousands of papers with 3 or 4 factor produced thousands of different coefficients (numbers) for these factors. Isn't it fuzzy?

     

    "So the whole academic community is wrong and you are right?"
    Probability that I'm right is too close to zero for somebody to care but as far as I know NOT whole academic community consider FF model or modern portfolio theory correct.
    IMO (again I'm not an expert in finance) a better job should be done to make finance more scientific. For example, econophysics is an attempt to improve understanding but as far as I can conclude from couple books I read their model isn't perfect and robust.
    Another attempt is behavioral finance. I definitely understand behavioral finance.less than econophysics but it seems /again from few books rather than from scholar papers which are prime source of any science/ that they are in the stage which any educated person can grasp. IMO, unfortunately, behavioral finance as in any social science faces with problems of representativeness and shift of base with time. I just wonder how folks in algorithmic trading (e.g. HFT) deal with behavioral finance.

     

    I do not feel good to argue with you before I read your book, please give me time and let's return to discussion after I do it.

     

    Sincerely
    SDS
    20 Feb 2012, 09:59 AM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    SDS
    I would agree, and it can happen due to bubbles, like when e/p was less than TIPS yield at height of internet related bubble.

     

    Have no clue what you are talking about in the second sentence.

     

    Like I said the FF model is used in VIRTUALLY all analysis of explanations of returns. YOu can conclude what you like from that.

     

    Finally I am big fan of behavioral finance, just wrote book on basically that subject, Investment Mistakes Even Smart People Make

     

    Best wishes
    Larry
    20 Feb 2012, 05:50 PM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Larry
    "Have no clue what you are talking about in the second sentence"
    Sorry probably my pure English. Let me rephrase
    Almost any curve can be fitted with 5 parameters model.
    In several papers coefficients for "size and value risks" within 3 factor FF model were calculated. Isn't it fuzzy that numerically these coefficients are different? I'd expect that a theory gives almost the same coefficient for all publications.

     

    Sincerely
    SDS
    20 Feb 2012, 11:17 PM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    All the model does is state that the exposure to the risk factors explains the vast majority of the difference in returns between diversified portfolios
    And remember it is not physics were there are laws. This is just a model of how the world works and like all models by definition it is not perfect.
    BTW-adding momentum as the fourth factor increases the explanatory power up into mid to high 90s. I would say it is pretty good model, though not perfect (:-))
    Best
    Larry
    21 Feb 2012, 10:36 AM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Larry
    Well we agree that a model is better word than a theory at least for this situation. I hope you agree that more than 1 model can describe the reality.
    SDS
    21 Feb 2012, 11:04 AM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    Of course, but question is which is best
    And you are clearly confusing two issues, MPT and the EMH are theories about how markets work, again not laws

     

    The FF model is not a theory, but a model that is used to explain returns.

     

    BTW-don't you think that with all the computer horsepower out there and with all the money devoted to the issue that if the FF model was not the best one we would have another better model and be using it? Note again Carhart came along and added the fourth factor which is now the standard model used, a four factor model. Perhaps someday we will have a fifth or sixth.
    Note we do have a two factor credit model, term and default risk.

     

    Best wishes
    Larry
    21 Feb 2012, 11:15 AM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Larry,

     

    I has no doubt that finance isn't physics and that finance numbers were chewed much more times than any scientific data. On another hand, I agree with quote (sorry don't remember author and exact words) "you cannot outperform if you do the same as others".

     

    "Carhart came along and added the fourth factor which is now the standard model used, a four factor model. Perhaps someday we will have a fifth or sixth."
    Yes, somebody will trade elegance for fitness as FF started but my point that this is not the way a good science is going. In science laws are axioms for a theory and a good theory (as well as a really good model) not only explains and but also PREDICTS with minimal number of axioms (assumptions).

     

    I know that I can make another model, I know that this model will be more elegant than e.g. FF92, I not sure that this model will be more precise that multifactor fitness, I know that with this model I might satisfy the above-mentioned quote, and I know that it takes significant effort to build a new model. Now if I estimate my personal return on effort I can see that it doesn't make sense for me to start such activity on my own.

     

    Sincerely
    SDS
    21 Feb 2012, 11:41 AM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    SDS
    That quote is simply wrong, Just have to define outperform. If you simply index, doing exactly what millions are doing you are GUARANTEED to outperform the vast majority of investors if yo ucan stay the course.

     

    Sorry, you have the second part wrong without question. Physics is about laws, certainties. Investing is about risk, if you could predict with certainty there would be no risk and no risk premiums. This seems to be point you get stuck on. At least from my readings of your thoughts.

     

    Best wishes
    Larry
    21 Feb 2012, 05:40 PM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Larry,
    I guess I need quote exactly because you misunderstand. Index perform as the market by definition, index-fees underperform. Quote is more or less in favor contrarians who as far as I know outperform market vs. indexers.
    IMO, risk is badly defined in finance. As far as I know I'm not along in this opinion.
    Physics is about behavior of natural and artificial objects and systems, so market is within its scope.
    SDS
    21 Feb 2012, 09:53 PM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    no evidence that contrarians outperform anything
    Again, risk cannot be defined as simply as you like with one figure, there are many types of risks and how the risks correlate matters also
    We have good definitions for example of SD, kurtosis, skewness, etc. But the problem is investing is all about UNCERTAINTY which is very different than risk. One is where you cannot measure the odds, the other is where you can. Then you have science. That seems to be point you miss.
    Best wishes
    Larry
    22 Feb 2012, 08:59 AM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Larry,
    "no evidence that contrarians outperform anything" - they claim so in books (Dreman, etc). Any serious analysis?

     

    Quantum physics is also about UNCERTAINTY, nobody use risk or another ill-defined term there.

     

    Well in this case risk cannot be axiom. It seems that you just don't understand this point - please my read
    Jean-Philippe Bouchaud, Economics needs a scientific revolution, Nature 455, 1181 (2008) available at
    http://bit.ly/yQ8Cos

     

    Sincerey
    SDS
    22 Feb 2012, 10:01 AM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    Dreman is not a "contrarian" though uses the term. He is simply a VALUE INVESTOR, buying high BtM, low P/E stocks and active "contrarians" don't outperform passive value funds.

     

    Best
    Larry
    22 Feb 2012, 11:46 AM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » Well are titles of Dreman's books wrong? I'd agree that contrarian and value strategies have huge overlap but in this case we should forget about any nuances in investment styles (which is not so bad for first approximation).
    "active "contrarians" don't outperform passive value funds"
    Examples, please.... Who are active contrarians? I think we are on the same page about any active stock investors: "less trading - better outcome". I can mathematically proof this statement differently than indexers did it but many traders don't care - they even decline invest in "new long-term US Treasure bond with 200% annual yield"! - see http://seekingalpha.co...
    22 Feb 2012, 01:09 PM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    Basically yes, people call them contrarian strategies so they can sell their skills, marketing hype, when it is nothing more than value strategy. You can see that in any loading analysis, bunch of BS.
    There are countless studies showing value active managers underperform. No different than any asset class.
    Larry

     

    BTW to show the paper you sent is basically wrong, using data from the Web site of Ken French, for the period 1928–2011, a high dividend strategy had an average annual return of 13.6 percent, compared to a 14.6 percent return of a U.S. large-cap value strategy. So much for that stuff.
    22 Feb 2012, 07:08 PM Reply Like
  • SDS (Seductive Dividend Sto...
    , contributor
    Comments (3335) | Send Message
     
    Author’s reply » For everybody:
    The article I sent Larry is "The quest for yield" by Elroy Dimson, Paul Marsh and Mike Staunton / the authors of excellent book "Triumph of the Optimists" (2002)/ appeared in Credit Suisse 2011 yearbook.

     

    Larry:
    I do not think that the paper is wrong at least facts are as solid as data-sources in the paper (I think they are the same as in their book). If Ken French has data ONLY for the period 1928–2011 the "London Trio" probably used another dataset for older results.

     

    "London Trio" result for components for US equity returns is the same as Societe Generale researches obtained (I guess from the same data source - Prof. R. Shiller Web site). I did reproduce Societe Generale result from Shiller's data.

     

    Sorry, but you have to accept the fact even if you dislike it. You might argue that the authors conclusion "Dividends have invariably been the largest component of real returns" isn't correct but you have to provide facts and coherent arguments why it is wrong.

     

    I do not remember that between 1900 and 2010 U.S. large-cap value strategy made 14.6% but anyway 13.6% at low risk (as they defined it) isn't bad.

     

    Another fact you will like that from their book result (shown in http://bit.ly/xFpiPl) I concluded that dividend growth strategy did not work in many countries before ~1950. Well, on another hand, the dividend growth strategy was quite robust and productive in USA for any non-overlapping 20years period from 1871 till 2011. I have 2 explanations why it happens but because one of explanation is non-fundamental, I cannot guaranty that the dividend growth strategy will be productive in next 50 years even in USA. Nevertheless, I'll stay as dividend investor till somebody shows me facts on significantly better strategy available or on fundamental flaw in dividends (MM paper isn't good argument).

     

    Sincerely
    SDS
    23 Feb 2012, 01:29 AM Reply Like
  • Larry Swedroe
    , contributor
    Comments (2405) | Send Message
     
    SDS
    It isn't that the divs were not the highest component. It is that the assumption if dividends were lower returns would have been lower that is implied. If divs were zero the returns would have been the same, basically.
    Stocks that have the same loading factors to beta, size and value have the same expected return regardless of their dividends. Dividends don't explain returns or we would need four factor model.

     

    Dividend growth strategies have not worked even since 1982 and I will show that in the second part of my series on dividend strategies tomorrow. Check out today's and tomorrows blog posts.

     

    I think or at least hope my two short pieces will demonstrate that, and remember divs strategies are more tax inefficient, and likely about to become even more so.

     

    Best wishes
    Larry
    23 Feb 2012, 09:56 AM Reply Like
Full index of posts »
Latest Followers

StockTalks

More »
Posts by Themes
Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.