Search found 323 matches

by camontgo
Thu Oct 11, 2012 3:54 pm
Forum: Investing - Theory, News & General
Topic: Technical Ken French data questions
Replies: 16
Views: 2927

Re: Technical Ken French data questions

Thanks for the replies. I don't understand why anyone really looks at "High minus Low" or "Small minus Big". This isn't how most people practically invest. Is there a good reason to look at the factors this way? In my mind, it's more intuitive and practical to think of, for example, the small value premium as the annual (or monthly) amount of return by which small value stocks have exceeded the total market. Thoughts? That's something I also wonder. SMB = 1/3 (Small Value + Small Neutral + Small Growth) - 1/3 (Big Value + Big Neutral + Big Growth). HML = 1/2 (Small Value + Big Value) - 1/2 (Small Growth + Big Growth) It seems to me that, because of the minus sign, if you want to capture the SmB you would have to not onl...
by camontgo
Thu Oct 11, 2012 3:41 pm
Forum: Investing - Theory, News & General
Topic: Technical Ken French data questions
Replies: 16
Views: 2927

Re: Technical Ken French data questions

SG SB SV BG? LB LV 1927 31.56 26.99 35.58 45.89 22.61 34.61 1928 31 40.45 42.35 46.66 31.84 25.02 1929 -45.31 -30.79 -36.81 -19.75 0.62 -4.41 1930 -35.27 -32.06 -45.34 -26.43 -29.4 -43.35 1931 -41.34 -48.69 -51.69 -36.38 -60.14 -57.89 1932 -5.29 -8.95 2.78 -7.32 -16.72 -4.31 Some of the data looks suspicious. Why does large blend perform worse than both large growth and large value in 1927? Why does it perform better than both in 1929? It should be in between. If the model were perfect, yes, but it's just a model. Over the short term random variations will cause these sorts of anomalies. Yes, also, due to the way the sorts work, the LV bucket has always had a relatively small number of stocks. So, perhaps it is especially prone to anomalou...
by camontgo
Thu Oct 11, 2012 1:25 pm
Forum: Investing - Theory, News & General
Topic: Are Large-Cap Stock Indexes Always Growth-Tilted?
Replies: 34
Views: 3992

Re: Are Large-Cap Stock Indexes Always Growth-Tilted?

However, back to my original question in the OP: Does the tilt to growth or value in the large cap indexes change over time? If so, wouldn't the value-factor loading depend upon the time period that one examined? Following up on some studies suggested by hafius500 above, I found these results for the S&P 500, over these specific time periods: I posted a graphic of the percent of TSM market cap which fell into each of Fama and French's six buckets (the ones they use to compute the HML and SMB factors) in 2011 here: http://www.bogleheads.org/forum/viewtopic.php?f=10&t=93095#p1508795 These percentages change over time. Here is the year by year data. Max in LG bucket was 79.1% in 2000, min in LG bucket was 30.9% in 1984. This data coul...
by camontgo
Wed Oct 10, 2012 8:27 pm
Forum: Investing - Theory, News & General
Topic: Technical Ken French data questions
Replies: 16
Views: 2927

Re: Technical Ken French data questions

Your calculations match what I get.

Here are my numbers for the geometric means. I converted the annual returns to log returns, averaged the log returns, and converted the average log return back to a simple return.

Image

Here are the arithmetic means for the same portfolios.

Image

I put some other tables here: http://www.bogleheads.org/forum/viewtop ... 5#p1508795
by camontgo
Wed Oct 10, 2012 1:34 pm
Forum: Investing - Theory, News & General
Topic: How to choose FF Factor Weightings
Replies: 46
Views: 14743

Re: How to choose FF Factor Weightings

Isn't that kind of self full-filling? What I mean is, if you take your universe of stocks and use it to calculate Beta, then a fund that holds every stock, it will have a Beta=1.00 and R^2 = 99.99. That's just definitional. In the same way, If you create a model based on sorting into certain bins, then funds that hold exactly those bins will match the model perfectly. So, if 3-factor regression is run on some SCV fund and it has R^2=97%, that may just be showing how closely the fund holdings match the stocks in the FF SVC bin. I think Cochrane does a good job addressing the "is the 3-factor model a tautology" question in Ch 20 of his Asset Pricing book. The relevant part of the book is available online here: http://faculty.chicag...
by camontgo
Tue Oct 09, 2012 4:30 pm
Forum: Investing - Theory, News & General
Topic: How to choose FF Factor Weightings
Replies: 46
Views: 14743

Re: How to choose FF Factor Weightings

It looks like they start with a 3,313 individual stocks. I assume they have the month end price and dividend for each stock over some period, which allows them to calculate the monthly total return for each stock. The data goes from 7/1926 to the present. (But most of the stocks did not exist for the whole period?) The total number of stocks changes each year. My table just captured 2011 as an example. There is a file called "6 Portfolios formed on Size and Book-to-Market" on the Ken French site which has all the details on the number of stocks used to form the portfolios for each year. It also has the monthly and annual returns for the 6 portfolios, so you could calculate HML and SMB from that file using the equations you have s...
by camontgo
Tue Oct 09, 2012 1:12 pm
Forum: Investing - Theory, News & General
Topic: How to choose FF Factor Weightings
Replies: 46
Views: 14743

Re: How to choose FF Factor Weightings

grayfox wrote:Annulized alpha = 0.059772 percent per year, close to zero
Beta = 0.56, less volatile than average stock
smb = -0.3236, tilt to larger stocks
hml = -0.027, close to zero =, so no value tilt
I ran this quick myself and I get almost the same results (I didn't make the dividend correction).

However, the returns in the script are decimal equivalent returns, not percents. So, 0.0598 = 5.98%. That alpha is quite large, though it is not statistically significant. (R^2 is low, so alpha is not well measured).
by camontgo
Tue Oct 09, 2012 10:24 am
Forum: Investing - Theory, News & General
Topic: How to choose FF Factor Weightings
Replies: 46
Views: 14743

Re: How to choose FF Factor Weightings

Description of Fama/French Benchmark Factors SMB = 1/3 (Small Value + Small Neutral + Small Growth) - 1/3 (Big Value + Big Neutral + Big Growth). HML = 1/2 (Small Value + Big Value) - 1/2 (Small Growth + Big Growth). I've got a few charts I created using the FF data...planning to someday use these in a blog post about calculating the factors. Here is how the stocks are divided. Note that the breakpoints are set using NYSE stocks only, then the other stocks are added to the portfolios after the size and value breakpoints are determined. http://www.calculatinginvestor.com/wp-content/uploads/2012/05/Portfolios-e1349795334287.png For example, size is divided into a top 50% and a bottom 50%, but it isn't really 50/50 by either number of stocks ...
by camontgo
Fri Oct 05, 2012 4:01 pm
Forum: Investing - Theory, News & General
Topic: Stock market vs interest rates
Replies: 4
Views: 1348

Re: Stock market vs interest rates

There is a good paper by Cliff Asness on the relationship between bond yields and stock yields ("Fight the Fed Model"), and he gives some examples which may relate to your questions. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=381480 Read his response to the various "arguments" on pages 12-13. ADDED: Here's my attempt at an explanation: Consider a simple Gordon Model for stock price: http://www.calculatinginvestor.com/wp-content/uploads/2012/10/gordon.gif P = Price D = Dividend g = expected dividend growth r_e = discount rate for equity, for simplicity assume this is the bond interest rate plus a constant ERP. The issue is that many economic factors which change interest rates also change expectations for g. Some ...
by camontgo
Fri Oct 05, 2012 2:31 pm
Forum: Investing - Theory, News & General
Topic: Kitces article on safe withdraw rates
Replies: 20
Views: 2841

Re: Kitces article on safe withdraw rates

I would take issue with the fact that the "margin of safety has decreased" Kitces himself writes that "low return environments are actually predictable from market valuation". If you believe that is even a little bit true (on average) then I think if follows that in a high valuation environment the margin of safety for a particular withdrawal rate is lower than it would be in a low valuation environment. If you don't believe that valuation has any utility for estimating future returns that's fine. However, in that case, do you unconditionally accept the "historical worst case" as a reasonable estimate for today's SWR? How do you know when you have enough historical data? Do you make any attempt to judge whethe...
by camontgo
Wed Oct 03, 2012 2:30 pm
Forum: Investing - Theory, News & General
Topic: Kitces article on safe withdraw rates
Replies: 20
Views: 2841

Re: Kitces article on safe withdraw rates

His logic seems wrong to me. He is saying that the recommended SWR is based on the historical worst-case scenario, and that scenario is pretty bad. So, even given our current low-bond-yield/high-stock-valuation environment, we aren't likely to do worse than that historical worst-case. I'd agree with that, but that's not all there is to it. As bond yields fall and stock valuations rise, the likelihood of doing even worse than the historical worst-case scenario rises..even if we think the probability is still small. The margin of safety has decreased, so rational investors may choose lower estimates of the forward looking SWR to get back to an acceptable margin of safety. Here's another way of looking at it. Using past data to predict the fut...
by camontgo
Fri Sep 28, 2012 10:27 pm
Forum: Investing - Theory, News & General
Topic: Comparison of small-cap value ETFs
Replies: 69
Views: 27637

Re: Comparison of small-cap value ETFs

Camontgo, are they more prone to big luck events like that (in either direction) than other indexes? Edit: I suspect that any index could have a lucky (or unlucky) rebalance, and they just happened to capture one of the largest momentum events in recent history. Now, a more significant concern is probably that the RAFI indexes were only created in, what, 2005? So any index returns before that would be biased to some extent. That's my guess too because the R^2 for PRF and PRFZ is similar to other large-value and small-value funds, so I don't think they are taking more non-systematic risk than other funds. If a fund's rebalancing methodology made them more prone to big luck events (positive or negative), then I would think that would show up...
by camontgo
Fri Sep 28, 2012 10:29 am
Forum: Investing - Theory, News & General
Topic: Comparison of small-cap value ETFs
Replies: 69
Views: 27637

Re: Comparison of small-cap value ETFs

Arnott has even stated himself that his funds got a "lucky" boost from their rebalancing on the 3rd Friday in March 2009. I would love to see two different data sets. One from PRFZ inception to current and one from 4/2010 to current. Although the latter is pretty short it would be interesting to see if it "got lucky" again and still generated positive alpha since that time. I've looked at this question for PRF. In the regressions, it is possible to just exclude the 12 month from 4/2009-3/2010 to see how this changes the results. I didn't have the MOM factor in my script and didn't have time to download from the Ken French site, so these are 3-factor only. The full sample 3-factor and full sample minus 4/2009-3/2010 are ...
by camontgo
Tue Sep 25, 2012 2:32 pm
Forum: Investing - Theory, News & General
Topic: History Doesn't Repeat...But It Rhymes!
Replies: 35
Views: 3602

Re: History Doesn't Repeat...But It Rhymes!

Portfolio's rebalanced annually: 1928-1990 (simulated period of Fama/French research) 100% Total Stock Index = +9.33% 18% TSM, 18% large value, 24% small value, 40% 5yr t-notes = +9.46% 1991-8/2012 (out of sample data) 100% Russell 3000 = +9.38% 18% Russell 3000, 18% Russell 1000VL, 24% Russell 2000VL, 40% 5yr t-notes = +9.52% :shock: I tried to look at the historical data in a different way. The history of "live" funds (both SCV and TSM) is limited, and the longer term academic datasets ignore some real world costs and constraints such as: 1. Academic datasets don't include any trading costs (i.e high bid-ask spreads for illiquid small and value stocks, also SCV funds must trade stocks more frequently). 2. Academic datasets don'...
by camontgo
Tue Sep 25, 2012 10:22 am
Forum: Investing - Theory, News & General
Topic: Sector vs. Company Diversification
Replies: 11
Views: 1395

Re: Sector vs. Company Diversification

Which of the hypothetical funds has the higher R^2 in the Fama-French model? (low R^2 = big residuals = lots of idiosyncratic risk) More generally, I would judge diversification by checking how well the funds in question track an index you believe is broadly diversified. For example, when comparing two U.S funds, you could see which tracks VTSMX more closely, or use the FF model and check R^2. If you are worried about country risk, I would measure the country's index against a Global Stock Market Index. There are some valid reasons to overweight your home country (taxes, hedging local consumption, etc). However, IMO, if the correlation with the global market is relatively low then concentrating a very large percentage of wealth in home coun...
by camontgo
Thu Sep 20, 2012 10:24 am
Forum: Investing - Theory, News & General
Topic: Will Your (TSM) Equity Bet Pay Off?
Replies: 222
Views: 15638

Re: Will Your (TSM) Equity Bet Pay Off?

A more efficient portfolio is one located closer to the northwest corner of a graph plotting return versus SD. Numerically one can use the Sharpe ratio. Many people will quickly comment that SD is not the only measure of risk, but it's a good starting point. IMO, This is really the key point for the whole argument. If the Sharpe ratio is 'more or less' the right metric to use then: 1. Tilting wins historically on risk-adjusted basis. The tilting strategy has a higher Sharpe ratio. 2. BUT!, it also looks like a free lunch. i.e. we get higher return with lower risk by overweighting small-value. Markets do not appear to be in an efficient equilibrium because investors can get better risk-adjusted returns by overweighting a sector of the marke...
by camontgo
Wed Sep 19, 2012 3:29 pm
Forum: Investing - Theory, News & General
Topic: Case Schiller question
Replies: 5
Views: 567

Re: Case Schiller question

I've looked at the methodology document in the past, and I don't think there is any inflation adjustment.

I think the link from Browser is based on someone doing the inflation adjustment on their own. Note that the index data from the S&P site is normalized to a value of 100 at the beginning of 2000, but the values shown in the multpl.com chart are much higher than 100 at the beginning of 2000.

Here is a link to the Case-Shiller methodology doc:

http://www.macromarkets.com/csi_housing ... ussion.pdf
by camontgo
Wed Sep 19, 2012 10:53 am
Forum: Investing - Theory, News & General
Topic: Will Your (TSM) Equity Bet Pay Off?
Replies: 222
Views: 15638

Re: Will Your (TSM) Equity Bet Pay Off?

He is talking about diversification and NOT best risk adjusted return, i.e. efficient frontier. Diversification is not the same as efficient frontier. The issue is the definition of risk-adjusted return. Yes, combining uncorrelated assets can lower standard deviation and potentially improve the Sharpe ratio. However, Fama doesn't believe that the average investor views portfolio optimization as a "mean-variance" optimization problem. If investors only cared about variance we would be in the one factor world of Markowitz's MPT and Sharpe's CAPM. The point of multi-factor models is that the definition of risk is more subtle. If you believe we are in a multi-factor world, the then classic Markowitz efficient frontier doesn't apply b...
by camontgo
Wed Sep 19, 2012 10:33 am
Forum: Investing - Theory, News & General
Topic: Will Your (TSM) Equity Bet Pay Off?
Replies: 222
Views: 15638

Re: Will Your (TSM) Equity Bet Pay Off?

He seems to be saying that the reason for tilting is "your tastes for these dimensions of risk." My interpretation: he means your speculative projections for the future of certain asset classes relative to the market. If he doesn't mean that, then what does he mean here by "taste?" The idea is that adding small and value tilts may increase expected return for a given amount of standard deviation...BUT, the tilted portfolio will have some other characteristics (higher correlation with "bad times") which are undesirable to many investors. The problem is that we don't understand the economic factors driving HML and SMB risk well enough to have good guidelines on who should overweight and who should underweight (I...
by camontgo
Tue Sep 18, 2012 3:46 pm
Forum: Investing - Theory, News & General
Topic: A question for value tilters about the average investor
Replies: 51
Views: 9752

Re: A question for value tilters about the average investor

If IJS (iShares S&P SmallCap 600 Value Index) has a PE of around 16 and the VTI (Vanguard Total Stock Market ETF) has a PE of around 14, then it would suggest to this dumb guy that 3 factor investors have already tilted the premium away by bidding up small-value. Shouldn’t a value stock index have a lower PE than the market? I think you are correct that small stock valuations are pricey right now, although there are also some distressed stocks in IJS with very low earnings. I'm not sure the PE of 16 is an "anomaly", but it is only part of the story. Here are all of the valuation numbers from Morningstar for IJS (SmallCap Value), IJT (SmallCap Growth), and VTI (TSM): IJS (SmallCap 600 Value Index ETF) Price/Prospective Earning...
by camontgo
Tue Sep 18, 2012 10:13 am
Forum: Investing - Theory, News & General
Topic: Will Your (TSM) Equity Bet Pay Off?
Replies: 222
Views: 15638

Re: Will Your (TSM) Equity Bet Pay Off?

What I observe is that many posters here do not seem to have kept up with the latest ideas of finance. They still toss around the term "equity risk premium" as if it is one number constant over time. They look at the historic long term excess return of stocks over bonds, or value over growth, or small over large, and then conclude that that is the risk premia going forward. That is an outdated idea, and is not consistent with current thought. Read Cochrane who summarizes it in some papers and presentations. Sure, Cochrane (and others) argues that risk-premia are time varying, but I've never heard him argue that there should be zero compensation for a legitimate risk...i.e. that a risk premium will go to zero. However, Cochrane is...
by camontgo
Mon Sep 17, 2012 8:28 pm
Forum: Personal Consumer Issues
Topic: Do you pick up pennies?
Replies: 56
Views: 4296

Re: Do you pick up pennies?

At the 4:45 mark in this video, Buffett says he'd pick up a dime. A penny would be a relative fortune to me :happy

http://www.youtube.com/watch?v=w4WTrIpTef8
by camontgo
Mon Sep 17, 2012 1:54 pm
Forum: Investing - Theory, News & General
Topic: Will Your (TSM) Equity Bet Pay Off?
Replies: 222
Views: 15638

Re: Will Your (TSM) Equity Bet Pay Off?

Fama says that the market portfolio is always on the efficient frontier and a perfectly acceptable mix of factors. He also says over-weighting small or value compared to market weights adds risk in an effort to get more return. More risk of course means the possibility of lower returns, perhaps dramatically so. :thumbsup The EMH academics who know this stuff best (Fama, French, Cochrane), believe that the market portfolio is an efficient choice. A portfolio with an SCV tilt may have higher expected returns (I believe it does), but it will also have higher risk. The OP implies that the average investor would be better off tilting, but if this is true then the market isn't in an efficient equilibrium. It is simply inconsistent to argue that ...
by camontgo
Fri Sep 14, 2012 9:40 am
Forum: Investing - Theory, News & General
Topic: Jason Hsu's Emerging FF Factors
Replies: 11
Views: 4188

Re: Jason Hsu's Emerging FF Factors

I don't know the answer to the specific questions you are raising, but, if you haven't read it already, I'd suggest reading the paper that FF wrote on factor models in developed markets. The Fama-French data for developed markets was used in the paper "Size, Value, and Momentum in International Stock Returns". Available here: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1720139 This paper focuses on developed markets rather than the emerging markets you are dealing with in your work, but it investigates the question of whether or not there is integrated pricing across regions. In other words, they compare local factor models for individual countries with regional asset pricing models (North America, Europe, Asia-Pacific). Th...
by camontgo
Thu Sep 06, 2012 1:19 pm
Forum: Personal Consumer Issues
Topic: What Book Are You Currently Reading? Part V
Replies: 3372
Views: 1565158

Re: What Book Are You Currently Reading? Part V

Just finished reading Again to Carthage by John L. Parker and listening to the audiobook Unbroken by Laura Hillenbrand.

Now reading Dark Pools by Scott Patterson.
by camontgo
Sat Aug 25, 2012 1:14 pm
Forum: Investing - Theory, News & General
Topic: How do you back test?
Replies: 26
Views: 3172

Re: How do you back test?

My 50% for 1 day is similar to 50% over time... Imagine a stock that always sells for $100 and pays 4% dividend (1% per quarter). Over 10 years, given that 0.99^40=0.669, Yahoo would adjusted prices would indicate you made (100-66.9)/66.9 = 49.5%. In reality, you would have made 40%. So, you would argue that the most accurate way to measure is to assume no reinvestment at all? The 40% is assuming all the dividends over 10 years are held for a 0% return? Is there any published source of total returns that computes things that way? I think any realistic total return number needs to consider reinvestment. You point out some good reasons why the assumption in the database calculation may differ from reality, but I'm not sure it is possible to ...
by camontgo
Sat Aug 25, 2012 10:30 am
Forum: Investing - Theory, News & General
Topic: How do you back test?
Replies: 26
Views: 3172

Re: How do you back test?

I used to use Yahoo Finance until I came across 2 issues: (1) Data IS inaccurate there. I found multiple mistakes in the data I got from it due to missing dividends from time to time (as I recall, in both indexes and individual stocks). (2) Computing returns based on adjusted price is simply wrong and leads to overall wrong results. To illustrate with an exaggerated example... . Agree with both, but the second one seems like it will be a very small error in practice. If you use a more typical $0.5 instead of $50, the error is only 0.0025%. Even if you compound an error of that size 100 times, you will only be off by a couple of tenths of a percent. In practice, the dividend will affect the price, especially if it is large, so even for a re...
by camontgo
Sat Aug 25, 2012 8:58 am
Forum: Investing - Theory, News & General
Topic: How do you back test?
Replies: 26
Views: 3172

Re: How do you back test?

Easiest: for any index that has an index fund, download the monthly prices from yahoo.com, and use the last column, the adjusted price, to derive the monthly returns. I took a stab at automating this method in Google Docs: https://docs.google.com/spreadsheet/ccc?key=0Am2C19KBrGAIdG1SZjctRkk2dmJ0UzJvTkRQTHFoa1E#gid=1 The return data is automatically downloaded from Yahoo! Finance when the ticker and start date are updated in the spreadsheet. The spreadsheet at the link above is "read-only", so a copy needs to be created if you want to change the target ticker and start dates. You then need to wait a minute or so for the spreadsheet to chug through the updates. You may occasionally get a "server error", but it should reco...
by camontgo
Sun Aug 19, 2012 8:51 am
Forum: Investing - Theory, News & General
Topic: Reference on Passive versus Active Performance?
Replies: 10
Views: 2481

Re: Reference on Index versus Active?

I think the best for showing data on investable funds in a clear, straighforward way is Bogle's (1992) "Selecting Equity Mutual Funds", but I haven't been able to find a public copy. These next two are available online, but are more technical than Bogle's; and not a straightforward comparison of investable funds since they evaluate active fund managers using factor models.....but, Carhart's study does a good job with the "persistence" issue. Carhart, Mark M., 1997, “On Persistence in Mutual Fund Performance,” Journal of Finance 52, 57-82. Fama, Eugene F. and Kenneth R. French, 2010, "Luck versus Skill in the Cross-Section of Mutual Fund Returns" Journal of Finance 65 Also, there is an Arnott, Berkin, and Ye stu...
by camontgo
Sun Aug 19, 2012 8:02 am
Forum: Investing - Theory, News & General
Topic: Any REAL bogleheads out there who are market weight int'l?
Replies: 20
Views: 2663

Re: Any REAL bogleheads out there who are market weight int'

Fama and French have a good discussion of international allocation here: http://www.dimensional.com/famafrench/2011/04/seeking-the-optimal-country-weighting-scheme.html#more Fama discusses under what conditions the global market weight portfolio would be considered "default" portfolio, and he gives several reasons why investors rationally choose to deviate from market weight. Fama's conclusion: In the end, faced with all the frictions in local and international investing, some simple advice for investment advisors may be best. The fundamental consideration is that the portfolio chosen should be highly diversified. If you think market prices are correct, you do not want risk that can be avoided through diversification. Various cons...
by camontgo
Fri Aug 17, 2012 2:20 pm
Forum: Personal Finance (Not Investing)
Topic: Google Spreadsheet Formula 4 Real Interest Rate?
Replies: 8
Views: 1773

Re: Google Spreadsheet Formula 4 Real Interest Rate?

You can get the real yields at various maturities from this site: http://www.treasury.gov/resource-center/data-chart-center/interest-rates/pages/textview.aspx?data=realyield There is an XML version of the data available here: http://www.treasury.gov/resource-center/data-chart-center/interest-rates/Datasets/real_yield.xml I'm not skilled with xpath queries (i.e. there may be a better way to do this), but you can pull the real yields into google docs with something like this (this example gives 30 year yields for each trading day of the current month): =importXml("http://www.treasury.gov/resource-center/data-chart-center/interest-rates/Datasets/real_yield.xml", "//TC_30YEAR") The last value in the list is the most recent 3...
by camontgo
Thu Aug 16, 2012 9:52 pm
Forum: Investing - Theory, News & General
Topic: Deleted (unless you want to discuss Apple)
Replies: 12
Views: 4542

Re: S&P 500 vs NASDAQ

I don't think the SPY returns shown in the chart are total returns...I'm not sure about the Nasdaq composite. Here are the S&P500 total returns for each year from1993-2011: 1993 10.08% 1994 1.32% 1995 37.58% 1996 22.96% 1997 33.36% 1998 28.58% 1999 21.04% 2000 -9.10% 2001 -11.89% 2002 -22.10% 2003 28.69% 2004 10.88% 2005 4.91% 2006 15.79% 2007 5.49% 2008 -37.00% 2009 26.46% 2010 15.06% 2011 2.05% That compounds to a 318% return...not including 2012 YTD. This is less than what you show for NASDAQ, but only a little (assuming NASDAQ returns are total returns) and the difference would about be reduced further with 2012 YTD returns for SPY included. The 1992 close for the S&P500 was 435.71 and today's was 1415.51, which is 225% gain. Th...
by camontgo
Thu Aug 16, 2012 3:57 pm
Forum: Investing - Theory, News & General
Topic: How deep a tilt? (sm, val, emerg, REIT, etc)
Replies: 56
Views: 5551

Re: How deep a tilt? (sm, val, emerg, REIT, etc)

It can't be rational for everyone to overweight an asset class (it isn't possible). So, whenever you think you are better off overweighting, then there must be someone who is either irrationally underweight the asset or is different enough from you that they are better off doing the opposite. In the case of concentrating your portfolio in X due to a large amount of X in the private sector, you're giving up diversification (by concentrating) due to the presence of uncapturable diversification. My question (to anyone) is: what's the mathematical or financial underpinning for this? Give me an example of how this can in some way improve my portfolio. In theory, it could be optimal for you to overweight the public shares of an asset class if the...
by camontgo
Thu Aug 16, 2012 1:58 pm
Forum: Investing - Theory, News & General
Topic: How deep a tilt? (sm, val, emerg, REIT, etc)
Replies: 56
Views: 5551

Re: How deep a tilt? (sm, val, emerg, REIT, etc)

Same with investing, there is nothing wrong with holding only one market dimension (just stocks/TSM-only). But we now operate in a world where we understand there are multiple sources of risk and priced (expected) returns. So by deciding not to include them, you are betting heavily that your preferred risk/return decision will work and work well enough for you to be successful. We of course know the perils of this approach, as the equity risk premium was dead from 65-81 (size and value were very large), small was dead from 82-99 (but market was huge and value was positive), from 95-99 value was very negative, size was negative, but the equity risk premium was 4X above average, and then from 00-12, the equity risk premium has been negative ...
by camontgo
Tue Aug 14, 2012 11:05 am
Forum: Investing - Theory, News & General
Topic: The important decisions in investing
Replies: 71
Views: 10678

Re: The important decisions in investing

rrppve wrote: I became a passive investor after my MBA and because of the concepts that were taught in finance including my Investing course using BKM.
I agree.

Prior to my MBA, I had read several popular books on indexing, and I even owned a couple index funds. However, I converted to a portfolio made up entirely of index funds only after taking an investment course which used BKM.
by camontgo
Tue Aug 14, 2012 9:47 am
Forum: Investing - Theory, News & General
Topic: a look at different passive strategies
Replies: 41
Views: 5023

Re: a look at different passive strategies

Larry, Shouldn't the table in the article show the FF3F alpha for each fund as well? My experience is that portfolios tilted towards small and value generally under-perform the FF3F model (they have meaningful negative alpha), and a larger factor loading on HML or SMB often comes at the cost of a larger negative alpha....beyond what can be explained by the expense ratio alone. This is expected to some degree since the factor portfolios don't include any trading costs, and very small stocks have higher bid-ask spreads and are costly to trade. However, the claim is often made that DFA manages these trading costs much more effectively that other fund companies (liquidity provider, etc), and we need the alphas to evaluate this claim and to do a...
by camontgo
Tue Aug 14, 2012 9:04 am
Forum: Personal Consumer Issues
Topic: What Book Are You Currently Reading? Part V
Replies: 3372
Views: 1565158

Re: What Book Are You Currently Reading? Part V

Just finished The New Geography of Jobs by Enrico Moretti and The Thousand Autumns of Jacob de Zoet (Audiobook) by David Mitchell

Now reading Retirement Portfolios by Michael Zwecher and listening to The Social Conquest of Earth by E.O. Wilson.
by camontgo
Sun Aug 12, 2012 10:41 am
Forum: Investing - Theory, News & General
Topic: Why isn't mean reversion easily visible in this chart?
Replies: 38
Views: 7117

Re: Where the momentum? Where's the RTM? Where's the beef?

The literature on momentum discusses relative performance among stocks. Stocks which have done well recently tend to outperform (for a short time) stocks which have done poorly recently. The effect is strong in historical data before considering costs. This is very different from a claim that market returns are autocorrelated. Even among individual stocks which exhibit momentum, autocorrelation is weak. I'm a skeptic on the net benefits of momentum after considering all costs (see comments here: http://www.bogleheads.org/forum/viewtopic.php?f=10&t=96085&hilit=+momentum#p1387958), but the effect that academics have documented is not the same as autocorrelation of the whole market. For example, here is the original paper on the moment...
by camontgo
Wed Aug 08, 2012 3:41 pm
Forum: Personal Investments
Topic: ESPP-Is it worth doing?
Replies: 20
Views: 3993

Re: ESPP-Is it worth doing?

Bob's not my name wrote:
camontgo wrote:If the stock goes up you can earn considerably more.
The OP has not stated that the plan has a lookback provision.
Right, I just re-read the OP. If there is no look-back then that comment doesn't apply.

My impression was that many ESPPs do have this provision, but I only have firsthand experience with one plan.
by camontgo
Wed Aug 08, 2012 3:26 pm
Forum: Personal Investments
Topic: ESPP-Is it worth doing?
Replies: 20
Views: 3993

Re: ESPP-Is it worth doing?

I think it is worth it if you are free to sell the stock shortly after the holding period ends. My company had a similar plan for about 11 years, and it was a valuable benefit (our plan was 15% discount with 6 month holding period). This program is like getting paid to hold a call option. If the stock goes down over the quarter you earn a minimum 5.26% on your contribution. If the stock goes up you can earn considerably more. There is some risk that the gain can be less than 5.26% if the stock falls during the short period between the end of the holding period and the time the shares are available to sell. In my experience, these small gains/losses averaged out over multiple holding periods. Here is a good article on flipping share in an ES...
by camontgo
Tue Aug 07, 2012 9:06 am
Forum: Personal Consumer Issues
Topic: What Book Are You Currently Reading? Part V
Replies: 3372
Views: 1565158

Re: What Book Are You Currently Reading? Part V

Just finished And a Bottle of Rum: A History of the New World in Ten Cocktails by Wayne Curtis. An entertaining history of rum and the New World. Now reading, The New Geography of Jobs, by Enrico Moretti.
by camontgo
Mon Aug 06, 2012 1:54 pm
Forum: Personal Consumer Issues
Topic: Wanting to become a faster reader
Replies: 26
Views: 3161

Re: Wanting to become a faster reader

This discussion reminds me of a Woody Allen quote:
I took a speed-reading course and read War and Peace in twenty minutes. It involves Russia. -Woody Allen
I've tried to teach myself some speed reading techniques, and I've found that I retain about as much as Woody when in speed reading mode.

My former boss seemed to be quite good at it though. He had taken a course when he was a student. I think it was the Evelyn Wood course.
by camontgo
Mon Aug 06, 2012 9:09 am
Forum: Investing - Theory, News & General
Topic: In theory, should TILT's price exceed its NAV?
Replies: 14
Views: 2279

Re: In theory, should TILT's price exceed its NAV?

A multi-factor risk model assumes that we don't live in a Markowitz style MPT world where investors only care about standard deviation as the measure of risk. If all we cared about was the Sharpe ratio of our portfolio, then it should be possible to show (as the OP is trying to do here) that the small and value premiums should not rationally persist. However, Fama and French believe that investors also care about covariance of returns with other factors in the economy. A tilted portfolio may not have higher standard deviation, but it is more likely to really fail badly when you care about performance most....that's the idea anyway. HML and SMB are priced risk factors because, all else equal, it is undesirable to have a portfolio with covari...
by camontgo
Fri Aug 03, 2012 1:13 pm
Forum: Investing - Theory, News & General
Topic: The most depressing financial newsletter ever
Replies: 38
Views: 6980

Re: The most depressing financial newsletter ever

"Club of Rome" type projections seem to be highly scientific, with lots of graphs and equations to undergird the predictions of The Final Days. Amazingly enough, the line charts and rows of numbers don't seem to make the predictions accurate. I agree. Rising prices create tremendous incentives for intelligent people to find innovative solutions. The cycle of incentives driving innovation has repeated throughout history. Grantham and the "Club of Rome" types sees a world with 7 billion mouths to feed, but others see 7 billion brains motivated to find solutions. Will we always be able to innovate our resource scarcity problems away? Maybe not. There will probably come a time when Jeremy Grantham, or his XXth century equiv...
by camontgo
Fri Aug 03, 2012 10:55 am
Forum: Investing - Theory, News & General
Topic: How do you back test?
Replies: 26
Views: 3172

Re: How do you back test?

BlueEars wrote: Here is the data from Yahoo for the end of July for VEU, first the daily and then the monthly:
Image

The (red arrow) monthly for July (shown as July 2) is actually the last day of July closing price. So if you ignore the erroneous day (should be July 31 but shows as July 2) and just look at the month, then the data is correct.
Thanks, this is good information. I didn't realize that the prices are end of the month since the dates shown are not. This makes my method for getting the monthly returns look a bit overly complex...although the result should be equivalent.
by camontgo
Thu Aug 02, 2012 3:02 pm
Forum: Investing - Theory, News & General
Topic: How do you back test?
Replies: 26
Views: 3172

Re: How do you back test?

Easiest: for any index that has an index fund, download the monthly prices from yahoo.com, and use the last column, the adjusted price, to derive the monthly returns. There are some inaccuracies, I've been told, but I've not encountered any. I like this method also. I use R (a free statistical software package) instead of Excel to download the adj closing prices and to calculate the monthly returns. R makes it easy to download data for many funds very quickly, but it requires being comfortable with some simple scripts. I have a demo of my method here: http://www.calculatinginvestor.com/2011/05/12/downloading-return-data/ Also, when calculating monthly returns from Yahoo prices, I start by downloading the daily adj price series rather than ...
by camontgo
Tue Jul 17, 2012 2:34 pm
Forum: Investing - Theory, News & General
Topic: forecasting the equity risk premium
Replies: 60
Views: 4850

Re: forecasting the equity risk premium

But people hate that advice. It sounds like giving up. It does not sell. It does not comfort people who want to know exactly what to do. I used to be like that. Then I got wiser as I got older. I probably am in the top 1% of folks who worked the math, I had access to vast databases of information, fancy mathematical software packages. But in the end it was a fools errand. Had you not worked the math to the extent you did, do you think you would have come to the same conclusion? I've worked the math quite a bit myself, and the paradox is that I've had to learn quite a lot in order to realize how little I know. I'm apparently not as far along on the path to wisdom though :happy ...I still think it is worthwhile to have some working estimate ...
by camontgo
Wed Jul 11, 2012 9:22 am
Forum: Personal Consumer Issues
Topic: Best Twenty Dollar Bourbon
Replies: 39
Views: 5407

Re: Best Twenty Dollar Bourbon

I usually splurge for a sipping whiskey in the $30-$40 range, but my father-in-law recently set-up a blind taste test for a few family members using several bourbons with price points up to $80 a bottle.

The Evan Williams Black Label beat out several more expensive bourbons. It wasn't the best, but several of us, myself included, liked it better than a couple of the bourbons in the $30-$40 range.

I've never purchased it myself, but I think it is even cheaper than your $20 target.
by camontgo
Mon Jul 09, 2012 4:21 pm
Forum: Investing - Theory, News & General
Topic: Optimal Investment Models Do Not Exist
Replies: 19
Views: 2833

Re: Optimal Investment Models Do Not Exist

Nice article. I agree that using financial models in a cookbook manner to precisely compute the optimal portfolio is a fool's errand. I do think models are very useful for shooting down bad arguments/ideas and for focusing the discussion/debate on the truly important questions. I've posted this quote before...it is one of my favorites from the introduction to Fama and Miller's 1972 Theory of Finance book: Readers will also note the absence of any detailed examples, of the kind often found in standard texts, purporting to show how to apply the theory in precise, quantitative terms, to real-world decision problems. This omission is less a move to save space than a reflection of our belief that the potential contribution of the theory of finan...
by camontgo
Mon Jul 09, 2012 11:17 am
Forum: Personal Consumer Issues
Topic: What Book Are You Currently Reading? Part V
Replies: 3372
Views: 1565158

Re: What Book Are You Currently Reading? Part V

I just finished The Blue Hour by Alonso Cueto.

Very good novel..especially if you are interested in history of Peru. The book won an award for best Spanish language novel back in 2005, but the English translation just came out on Kindle about a month ago. It was initially available on Kindle only, but I just checked Amazon and looks like the print version is available starting today.

Now starting Fortune's Formula by William Poundstone. I've been reading about the Kelly Criterion based on some previous discussions on this forum, and, in the process, I came across several recommendations for this book. I am enjoying it so far.