Playing with Callan's periodic tables of investment returns

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
FIREchief
Posts: 6588
Joined: Fri Aug 19, 2016 6:40 pm

Re: Playing with Callan's periodic tables of investment returns

Post by FIREchief »

LadyGeek wrote: Tue Sep 04, 2018 6:52 pm From the perspective of a new investor, that statement is confusing. I don't see the distinction between (a) and (b).
It's probably made more clear by looking at an actual Callan table of the major equity catetories. It is often shown by (let's assume) well intentioned advisors to clients in an effort to extol the virtues of broad diversification. This is how I first saw one. You focus on the top row and as you go accross you might see "emerging markets," "EAFE", etc. pop up and then disappear. The message is that by investing in all the categories, you'll be assured of having something invested in the "winner" each year. If you then look at the bottom couple of rows, you'll tend to see the same ones pop up and disappear. It is when you find the right color for S&P 500 or US Total Market that you notice they are never at the top or the bottom, but tend to bounce around the middle. Some investors see this as an argument to own everything in the world. Others would see it as an argument to just go with large cap or total US. We'll agree on an answer to that just as soon as our threads on international stocks go away.... :mrgreen:
I am not a lawyer, accountant or financial advisor. Any advice or suggestions that I may provide shall be considered for entertainment purposes only.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

LadyGeek wrote: Tue Sep 04, 2018 6:52 pm Have you experimented with a stacked bar (column) chart? It would be similar to the above, but the height of each block would be scaled relative to the total. I think this will give you a sense on the magnitude of the relative returns (larger returns have larger block sizes), which may be closer to heyyou's suggestion.
Yes, I tried and wasn't too convinced by the result, as I mentioned in this post. The height of a block would have to be relative to the height of the next one, making some slices really thin. And the visual outcome just didn't seem to convey as powerful a message as the regular square blocks.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

LadyGeek wrote: Tue Sep 04, 2018 6:52 pm I would agree, but should that post be in the wiki?
I would vote 'no'. It isn't always true that a total-something (stock or bonds) hovers in the middle. I think this would just create confusion.
LadyGeek wrote: Tue Sep 04, 2018 6:52 pm On a more technical note, footnote 4 describes a metric known as "coefficient of variation". Would that be useful here?
I added some basic stats in the spreadsheet (e.g. nominal/real CAGR, standard-deviation). The CV metric is something that Barry Barnitz seems be fond of, it is kind of a poor's man Sharpe ratio, but as far as I know, this isn't a very popular financial metric. We don't have it in Simba, where there is a much richer set of metrics. I am not too keen to add it, unless there is a clear demand for it. If somebody is interested, they could add a row to compute it themselves, this is trivial to do.

I have been also wondering about adding the dispersion chart we have in the Wiki. It seems more directly related to the Callan's table, and would provide a sense of scaling. This would seem like a nice addition, why not.
User avatar
LadyGeek
Site Admin
Posts: 71307
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Playing with Callan's periodic tables of investment returns

Post by LadyGeek »

^^^ I agree on both points.

- 'No' to providing a perspective that can be misinterpreted by new investors.

- The "coefficient of variation" did seem to be a variation of the Sharpe ratio, but from a different perspective. Too many metrics can be counter-productive.
siamond wrote: Tue Sep 04, 2018 10:03 pm
LadyGeek wrote: Tue Sep 04, 2018 6:52 pm Have you experimented with a stacked bar (column) chart? It would be similar to the above, but the height of each block would be scaled relative to the total. I think this will give you a sense on the magnitude of the relative returns (larger returns have larger block sizes), which may be closer to heyyou's suggestion.
Yes, I tried and wasn't too convinced by the result, as I mentioned in this post. The height of a block would have to be relative to the height of the next one, making some slices really thin. And the visual outcome just didn't seem to convey as powerful a message as the regular square blocks.
Sorry, missed that.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

staythecourse wrote: Tue Sep 04, 2018 7:17 pmCan the periodic table reflect more of all the MAJOR subasset classes and not the granular country or cap size subassets? I think a quilt of: sp500, russell 2000v, EAFE, EM, TIPS, TBM, Tinternational bonds, cash, gold, commodities (CCF), REITS, etc... would be the most useful represented over the last 20 years (for example). Then an interactive way so one can make one's on portfolio weights and see how it shows up in that same quilt over the last 20 years.
This thread is about a generic spreadsheet allowing to create whatever Callan's table you want. This being said, to ease the life of those who like predefined formats, I added a few predefined sets of portfolios, including the one mapping the best to the 'classic' table provide on the Callan's Web site.

Image

If you're interested by other categories (e.g. REITs, etc), please go back to the first post of this thread, follow the instructions to download the spreadsheet, and don't hesitate to customize it at will...
User avatar
FreeAtLast
Posts: 721
Joined: Tue Nov 04, 2014 9:08 pm

Re: Playing with Callan's periodic tables of investment returns

Post by FreeAtLast »

siamond wrote: Tue Sep 04, 2018 8:32 am Just for fun, I assembled a table for bonds. Here it is. Click to see a larger image.

LTT = Long-Term Treasuries; ITT = Intermediate Term Treasuries; STT = Short Term Treasuries; T-Bills = Treasury Bills; IT Corp = Intermediate Term Corporates; HY Corp = High Yield Term Corporates; TIPS = Treasury Inflation-Adjusted Securities; Global = Global (unhedged) bonds; Int'l = ex-US (hedged) bonds; TBM = US Total Bonds Market.

Image
The "Callan Table" for bonds is an absolutely brilliant idea. Put it in the Wiki!
Illegitimi non carborundum.
User avatar
FIREchief
Posts: 6588
Joined: Fri Aug 19, 2016 6:40 pm

Re: Playing with Callan's periodic tables of investment returns

Post by FIREchief »

I don't believe that anybody has suggested that that post should be included in the wiki. I'm the one who made it, and I certainly didn't! 8-)
I am not a lawyer, accountant or financial advisor. Any advice or suggestions that I may provide shall be considered for entertainment purposes only.
staythecourse
Posts: 6993
Joined: Mon Jan 03, 2011 9:40 am

Re: Playing with Callan's periodic tables of investment returns

Post by staythecourse »

siamond wrote: Tue Sep 04, 2018 10:23 pm
staythecourse wrote: Tue Sep 04, 2018 7:17 pmCan the periodic table reflect more of all the MAJOR subasset classes and not the granular country or cap size subassets? I think a quilt of: sp500, russell 2000v, EAFE, EM, TIPS, TBM, Tinternational bonds, cash, gold, commodities (CCF), REITS, etc... would be the most useful represented over the last 20 years (for example). Then an interactive way so one can make one's on portfolio weights and see how it shows up in that same quilt over the last 20 years.
This thread is about a generic spreadsheet allowing to create whatever Callan's table you want. This being said, to ease the life of those who like predefined formats, I added a few predefined sets of portfolios, including the one mapping the best to the 'classic' table provide on the Callan's Web site.

Image

If you're interested by other categories (e.g. REITs, etc), please go back to the first post of this thread, follow the instructions to download the spreadsheet, and don't hesitate to customize it at will...
Much thanks. Is there a way to make a portfolio and see where it falls into that quilt?

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

staythecourse wrote: Wed Sep 05, 2018 9:09 am
Much thanks. Is there a way to make a portfolio and see where it falls into that quilt?
Yes, you can define and compare portfolios made of multiple asset classes. Give it a try!
livesoft
Posts: 75914
Joined: Thu Mar 01, 2007 8:00 pm

Re: Playing with Callan's periodic tables of investment returns

Post by livesoft »

Aren't you expecting a Cease & Desist Letter from those Callan folks for using their name?

You had better quickly trademark these Siamond Tables before someone else tries to sequester them.
Wiki This signature message sponsored by sscritic: Learn to fish.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

^^^^
I'd rather receive a check for promoting the name... In all fairness, all credits to the Callan folks for creating this type of representation, this was a brilliant idea. I hope they won't mind getting more exposure.
User avatar
LadyGeek
Site Admin
Posts: 71307
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Playing with Callan's periodic tables of investment returns

Post by LadyGeek »

Remember that the official "Callan Periodic Table"is copyright by Callan LLC and is in the wiki as " Reprinted with permission." (Yes, I really do have an email from Callan giving permission to reproduce the table.)

Callan is not the only company to do this, however. A "generic" periodic table is fine. For example - "Siamond's periodic table of annual bond returns."

Readers should note that our sister Canadian site's wiki has the Periodic table of annual returns - finiki, the Canadian financial wiki
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
DoTheMath
Posts: 456
Joined: Sat Jul 04, 2015 1:11 pm
Location: The Plains

Re: Playing with Callan's periodic tables of investment returns

Post by DoTheMath »

siamond wrote: Sat Sep 01, 2018 12:13 pm Here is a packaging of my Callan Table Calculator spreadsheet. Please download it here, and use a reasonably recent version of Excel:
https://drive.google.com/uc?id=1-EZIwGV ... t=download

The 'README' tab provides instructions. Those of you used to the Simba backtesting spreadsheet will find a familiar structure, allowing the user to name and define 10 portfolios, using some controls to be more specific about exactly what type of return (e.g. nominal, real, premium over a benchmark, annual or time intervals) you'd like to see in the Callan's table. I provided two formats for Callan tables, the regular 'quilt' and the distribution around a 'zero' axis.

Feedback welcome...

PS. after solving a small compatibility issue, it appears to work fine in LibreOffice, although I didn't play much with it.
Thanks! I look forward to playing with it.
“I am losing precious days. I am degenerating into a machine for making money. I am learning nothing in this trivial world of men. I must break away and get out into the mountains...” -- John Muir
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

About the wiki page... It starts by the following statement that Taylor issued back in 2012:

The Callan Table is the best visual information showing the importance of diversification, reversion-to-the-mean, and the impossibility of forecasting asset-class returns. It is a primary reason Bogleheads favor total market index funds.

Although I agree with the general thrust of the statement, a couple of things do not seem quite right.
- I don't quite see that this type of table clearly illustrates reversion-to-the-mean properties (Telltale charts do a much better job of showing such property - when it existed).
- Also, the forecasting thing is true in the short-term, and well illustrated by such table, I agree, but the statement is inappropriately broad (and could be viewed as contradicting expected returns longer-term models promoted by Jack Bogle, Dr. Bernstein and others).
- Finally, there is an large part of the Bogleheads community which isn't only using total market funds (e.g. slice-and-dice folks)

I would suggest to rephrase:

The Callan Periodic Table is the best visual information showing the importance of diversification and the futility of short-term forecasting. It is a primary reason for which many Bogleheads favor total market index funds.

And then change the following text in the first section accordingly. Also, there are 10 asset classes in the latest Callan's table (the one from the Callan folks!), not 9. We should make the text more generic, no point mentioning a precise number of rows.

Agreed?
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

Based on the last few posts, I issued a new revision of the spreadsheet.

Rev1c
1. Settled on 'periodic table' terminology when referring to this type of table; renamed spreadsheet file accordingly
2. Added a dispersion graph in Define_Portfolios

Same download link:
https://drive.google.com/uc?id=1-EZIwGV ... t=download
User avatar
LadyGeek
Site Admin
Posts: 71307
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Playing with Callan's periodic tables of investment returns

Post by LadyGeek »

siamond wrote: Wed Sep 05, 2018 7:52 pm About the wiki page... It starts by the following statement that Taylor issued back in 2012:

The Callan Table is the best visual information showing the importance of diversification, reversion-to-the-mean, and the impossibility of forecasting asset-class returns. It is a primary reason Bogleheads favor total market index funds.

Although I agree with the general thrust of the statement, a couple of things do not seem quite right.
- I don't quite see that this type of table clearly illustrates reversion-to-the-mean properties (Telltale charts do a much better job of showing such property - when it existed).
- Also, the forecasting thing is true in the short-term, and well illustrated by such table, I agree, but the statement is inappropriately broad (and could be viewed as contradicting expected returns longer-term models promoted by Jack Bogle, Dr. Bernstein and others).
- Finally, there is an large part of the Bogleheads community which isn't only using total market funds (e.g. slice-and-dice folks)

I would suggest to rephrase:

The Callan Periodic Table is the best visual information showing the importance of diversification and the futility of short-term forecasting. It is a primary reason for which many Bogleheads favor total market index funds.

And then change the following text in the first section accordingly. Also, there are 10 asset classes in the latest Callan's table (the one from the Callan folks!), not 9. We should make the text more generic, no point mentioning a precise number of rows.

Agreed?
Not quite. First, I corrected the number of asset classes (from 9 to 10) and revised the citation note accordingly. See: Callan periodic table of investment returns

Reversion to the mean - I do see a vague trend towards reversion to the mean. Focus on a single color and notice how the returns vary from high to low over time. The "average" tends towards zero. Crude, yes. However, I think it makes the point. Considering this article is focused on the "very new" investors, I think simplicity is important.

Expected returns - I think the point of the table is intended to contradict the long-term models. "Past performance does not predict future performance" is clear and visually presented.

Bogleheads community sample size - Your observations are biased towards members who post in discussions similar to this one, i.e. experienced investors who choose to deviate from the total market approach. The article is intended for the many, many new investors who don't understand this stuff and should follow the "ideal" approach. The Callan table gets them started off on the right foot. If they want to change their approach later - and understand the increase in risk due to a tilted portfolio - that's fine.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
LadyGeek
Site Admin
Posts: 71307
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Playing with Callan's periodic tables of investment returns

Post by LadyGeek »

siamond wrote: Wed Sep 05, 2018 8:37 pm Based on the last few posts, I issued a new revision of the spreadsheet.

Rev1c
1. Settled on 'periodic table' terminology when referring to this type of table; renamed spreadsheet file accordingly
2. Added a dispersion graph in Define_Portfolios

Same download link:
https://drive.google.com/uc?id=1-EZIwGV ... t=download
The URL you've provided is a direct download. Can you supply a link which allows one to view the spreadsheet from Google Drive? This will allow the spreadsheet to be viewed directly in the wiki (if it's added later...).

Minor typo: README!A22 - "slighty" should be "slightly".

It looks fine in LibreOffice Calc.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

LadyGeek wrote: Wed Sep 05, 2018 8:58 pmThe URL you've provided is a direct download. Can you supply a link which allows one to view the spreadsheet from Google Drive? This will allow the spreadsheet to be viewed directly in the wiki (if it's added later...).
I could, but I'd rather not. I do the same with the Simba links. First, this saves a download step for the end user. Next, the spreadsheet doesn't work properly with Google Sheet, I tried...
LadyGeek wrote: Wed Sep 05, 2018 8:58 pmMinor typo: README!A22 - "slighty" should be "slightly".
Thanks, good catch, fixed.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

LadyGeek wrote: Wed Sep 05, 2018 8:47 pm
siamond wrote: Wed Sep 05, 2018 7:52 pm About the wiki page... It starts by the following statement that Taylor issued back in 2012:

The Callan Table is the best visual information showing the importance of diversification, reversion-to-the-mean, and the impossibility of forecasting asset-class returns. It is a primary reason Bogleheads favor total market index funds.

Although I agree with the general thrust of the statement, a couple of things do not seem quite right.
- I don't quite see that this type of table clearly illustrates reversion-to-the-mean properties (Telltale charts do a much better job of showing such property - when it existed).
- Also, the forecasting thing is true in the short-term, and well illustrated by such table, I agree, but the statement is inappropriately broad (and could be viewed as contradicting expected returns longer-term models promoted by Jack Bogle, Dr. Bernstein and others).
- Finally, there is an large part of the Bogleheads community which isn't only using total market funds (e.g. slice-and-dice folks)

I would suggest to rephrase:

The Callan Periodic Table is the best visual information showing the importance of diversification and the futility of short-term forecasting. It is a primary reason for which many Bogleheads favor total market index funds.

And then change the following text in the first section accordingly. Also, there are 10 asset classes in the latest Callan's table (the one from the Callan folks!), not 9. We should make the text more generic, no point mentioning a precise number of rows.

Agreed?
Not quite. First, I corrected the number of asset classes (from 9 to 10) and revised the citation note accordingly. See: Callan periodic table of investment returns

Reversion to the mean - I do see a vague trend towards reversion to the mean. Focus on a single color and notice how the returns vary from high to low over time. The "average" tends towards zero. Crude, yes. However, I think it makes the point. Considering this article is focused on the "very new" investors, I think simplicity is important.

Expected returns - I think the point of the table is intended to contradict the long-term models. "Past performance does not predict future performance" is clear and visually presented.

Bogleheads community sample size - Your observations are biased towards members who post in discussions similar to this one, i.e. experienced investors who choose to deviate from the total market approach. The article is intended for the many, many new investors who don't understand this stuff and should follow the "ideal" approach. The Callan table gets them started off on the right foot. If they want to change their approach later - and understand the increase in risk due to a tilted portfolio - that's fine.
Er. No. Trying to help new investors isn't an excuse to make factually wrong statements. And I really do not see how my proposed change would confuse any new investor. It is simply more accurate, period. Please show me otherwise?

Reversion to the mean: well, afaik, there is no known historical data series over a period of 20 years that shows a return-to-the-mean property in a truly convincing manner. Please check the telltale charts in this blog entry, and you'll see what I mean. Also, the fact that such periodic tables do not capture the scale of numbers make them especially inadequate to illustrate any RTM property. The Callan table is great, but NOT for that, this is too short of a time horizon, and simply not the right graphic.

Past/future/returns/etc: the evidence about expected returns models is fairly strong (I can bore you with many charts and numbers if you insist!!), and the "past performance doesn't predict blah" statement keeps being misused, it was never intended for anything else than short-term predictions for active funds. And a typical periodic table is pretty short-term. Yes, I am keenly aware that a significant faction of Bogleheads just doesn't buy into the expected returns models (usually missing the point about its statistical nature), but there are quite a few who do, starting by John Bogle himself. In a case like that where there is no consensus, it seems to me that we'd better compromise with a statement that is acknowledged as correct by all parties. The existing statement is NOT like that. I think my proposed change would be. Am I wrong?
User avatar
LadyGeek
Site Admin
Posts: 71307
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Playing with Callan's periodic tables of investment returns

Post by LadyGeek »

No, you're not wrong. Thinking harder, you make a good point. I'm looking at this wiki page: Mean reversion
...returns can be very unstable in the short run but very stable in the long run.
Considering everything you've stated, I am now agreeing that the wiki page should be updated.

We should get a consensus on the wording from a few more members.

===============
I think it boils down to the time horizon. The Callan charts span 10 years, while the other analyses (Bogle, etc.) go much longer.

One of the most important considerations for deciding on your investments is your time horizon. My concern is that investors will see "futility of short-term forecasting" and then wonder "what is long-term forecasting"? For someone with a 30 year horizon, your statement would apply.

For someone with a 5 year horizon, there is concern for sequence of return risk - what does this chart tell you?

Perhaps shortening your statement to remove the time-horizon concerns would be appropriate:

The Callan Periodic Table is the best visual information showing the importance of diversification. It is a primary reason for which many Bogleheads favor total market index funds.

would make more sense from a general perspective. This is the how the article is currently worded. So, perhaps I've provided some rationale to keep the original wording for the diversification part.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

Ladygeek, thank you for listening to my side of the argument.
LadyGeek wrote: Wed Sep 05, 2018 8:47 pmReversion to the mean - I do see a vague trend towards reversion to the mean.
Not to beat a dead horse, but let me just add one more thing about return-to-the-mean (after sleeping on it!). Take the example of Small-Cap Value (SCV). When looking at the factors-oriented chart I posted in the OP, we see SCV in pretty random places in the ranking. One could get a sense of return-to-the-mean (i.e. not make a difference with Total-Market -TSM- in the big picture). But this is factually wrong (or at least it has been so far). In the corresponding time period, when looking at a Telltale chart, SCV actually diverged from TSM quite significantly (i.e. it returned 80% more). Which we don't perceive on the periodic table because such representation mutes the scaling of the annual returns. Now, one could argue that the time period is too short for properly observing RTM, and that the pendulum would (or could) reverse itself. Fair point. It turns out that when taking a broader time horizon (see the Telltale chart in the blog I referred to), SCV kept diverging more and more. But for other asset classes (e.g. Small-Cap Blend), occasional divergence over the course of a couple of decades did happen, while the long term evidence is much more debatable, and RTM (relative to TSM) proponents may very well have a point in this respect.

Sorry if I'm going in the weeds here, but my point is that the intuitive RTM perception (the 'vague trend' you observed) that one might develop by looking at a periodic table can be quite misguided and factually wrong, as a side-effect of both lack of scaling and relatively short time period. Hence my strong inclination to eliminate any RTM consideration from the corresponding wiki page.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

LadyGeek wrote: Wed Sep 05, 2018 10:53 pmI think it boils down to the time horizon. The Callan charts span 10 years, while the other analyses (Bogle, etc.) go much longer.

One of the most important considerations for deciding on your investments is your time horizon. My concern is that investors will see "futility of short-term forecasting" and then wonder "what is long-term forecasting"? For someone with a 30 year horizon, your statement would apply.

For someone with a 5 year horizon, there is concern for sequence of return risk - what does this chart tell you?

Perhaps shortening your statement to remove the time-horizon concerns would be appropriate:

The Callan Periodic Table is the best visual information showing the importance of diversification. It is a primary reason for which many Bogleheads favor total market index funds.

would make more sense from a general perspective. This is the how the article is currently worded. So, perhaps I've provided some rationale to keep the original wording for the diversification part.
LOL, you're going pretty far in the other direction now, but... I have a sense that we're losing something important. I agree that my 'short-term forecasting' wording might be opening a can of worms. Let me suggest to come at it from another angle. I would suggest to rephrase:

The Callan Periodic Table is the best visual information showing the importance of diversification and the futility of market timing. It is a primary reason for which many Bogleheads favor total market index funds.

or maybe even:

The Callan Periodic Table is the best visual information showing the importance of diversification and the futility of market timing. It is a primary reason for which many Bogleheads favor total market index funds, while electing to stay the course.

And then adjust the following text in the wiki article to refer to the market timing wiki page instead of the (somewhat debatable) "Past performance does not blah" statement.

This indirectly captures my point because market timing is intrinsically a short-term activity. This also keeps the essence of what Taylor was stating, I believe, while putting more focus on something actionable (do NOT market-time, DO stay the course) than forecasting or not (forecasting isn't an end in itself). And this keeps things simple, intuitive, and easily understandable for new investors. What do you think?
User avatar
LadyGeek
Site Admin
Posts: 71307
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Playing with Callan's periodic tables of investment returns

Post by LadyGeek »

siamond wrote: Thu Sep 06, 2018 6:34 am LOL, you're going pretty far in the other direction now, but... I have a sense that we're losing something important. I agree that my 'short-term forecasting' wording might be opening a can of worms. Let me suggest to come at it from another angle. I would suggest to rephrase:...
Your first statement is simple and to the point. Adding extra phrasing, such as "while electing to stay the course", will open a different can of worms.

First, here's the original text:
The Callan Table is the best visual information showing the importance of diversification, reversion-to-the-mean, and the impossibility of forecasting asset-class returns. It is a primary reason Bogleheads favor total market index funds.
And noting:
siamond wrote: Thu Sep 06, 2018 6:34 am And then adjust the following text in the wiki article to refer to the market timing wiki page instead of the (somewhat debatable) "Past performance does not blah" statement.
The Market timing page does not convey a clear message. It's an accurate definition, but it does not help new investors. I would suggest instead to point to the Bogleheads' investment philosophy page - which is intended for new investors and conveys the appropriate message.

I propose this text:
The Callan Periodic Table is the best visual information showing the importance of diversification and the futility of market timing. It is a primary reason for which many Bogleheads favor total market index funds.
Regarding the "following text" and "Past performance does not predict future performance":
  • Past performance does not predict future performance.
  • Diversification: By owning the entire market (all of the asset classes), susceptibility to changes in market variations is minimized.
  • Reversion-to-the-mean: Large variations over a short period of time, but tends to be stable when viewed over the long term.[note 4]
It is debatable. But, I don't think there is any other way to convey this message clearly and succinctly. I think the 3rd point explains the difference between short-term and long-term investing. Footnote 4 is a technical deep-dive for those who want to see the "proof".

My recommendation would be to not change this text.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

^^^^^^^^^^ (the Ladygeek trademarked quotation style!)

Slowly converging... :wink:

I am fine with the simpler top statement. Agreed.

Pointing to the Bogleheads philosophy page, ok, why not. I'd argue that if the market timing page isn't clear, it should be improved, but ok, one thing at a time. Agreed.

I am reluctantly ok with leaving the "past performance... blah" bullet point. It is a misleading statement when taken too literally, but we have it all over the place, and I agree that it is a good starting point, illustrated by the periodic table and something useful for new investors. Agreed.

I am strongly against keeping the bullet point about return-to-the-mean. As I explained in this post, periodic tables are actually VERY misleading when it comes to this topic, due to lack of scaling and lack of deep history. I don't see that Note 4 proves anything, it's just a bunch of diverse numbers for a given time period. Leaving this statement is simply factually wrong and misleading. And if you think that the market timing page is confusing, well, the mean reversion page isn't exactly clear for newbies (it confuses me, to be honest!). NOT a good thing for new investors. Plus this would be inconsistent with the revised top statement.

Could we please compromise on what we agree on here? And let's move on. :wink:
User avatar
LadyGeek
Site Admin
Posts: 71307
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Playing with Callan's periodic tables of investment returns

Post by LadyGeek »

When in doubt, go to the source. Here's the thread which contains Taylor's post: Callan Periodic Table of Investment Returns. Although Taylor mentions reversion-to-the-mean and the impossibility of forecasting asset-class returns, the ensuing discussion does not mention this.

( I then did a quick scan of the referenced publications and fixed a few broken article links :) ).

I could not find any mention on the topic of reversion-to-the-mean and agree to remove the 3rd bullet.

However, that would also remove the note containing the coefficient of variation. I'd like to keep that concept. Could the bullet be moved to be part of that note, then cite the note in a different context?

================
Page 3 of The Importance of Diversification by Allianz states in the footnote "Past performance is no guarantee of future results." Which I am using to support a point we already agreed upon - to keep the "Past performance does not predict future performance." bullet.

===============
We should further clarify that the table refers to asset classes - not individual stocks. The referenced publications and forum threads are clear on this point. I propose this revised text:
The Callan Periodic Table is the best visual information showing the importance of asset class diversification and the futility of market timing. It is a primary reason for which many Bogleheads favor total market index funds.
We are mostly in agreement and just need to figure out where to put the revised note.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

Adding 'asset class' verbiage: yes, good point.

Keeping note 4: I am not very attached to the CV metric per se, but showing the full table of stats is indeed useful. I would also suggest that note 3 is misplaced, it doesn't illustrate the sentence to which it is currently associated. Both notes (the graph and the metrics) are very useful in showing the periodic table in perspective, both scaling and over time.

Maybe we should add a new section warning users that a simple ranking can be a bit misleading, as was discussed in this very thread (and previous ones). And using the dispersion graph and the stats table to illustrate the point. I basically did the same in the spreadsheet.

Why don't we add a new section, something like '"Putting Callan's Periodic Table in perspective":

Periodic tables provide a great visual about diversification benefits, but tend to be more qualitative than quantitative. The simple ranking from worse to best notably does not allow to easily appreciate the scaling of annual returns. The following dispersion graph is therefore useful to put such periodic table in perspective.

<< insert graph >>

In addition, it is challenging to get a sense of returns averaged over a period of time with a periodic table. The following table of statistics is therefore useful to consider.

<< insert the multi-tab stats table >>


Then I would suggest to keep the CV explanation as a footnote, associated with the corresponding cell text in the table.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

PS. hm, the stats table isn't full up to date. Only 9 columns in there, not 10. The High Yield bonds are missing. Another thing to fix.
User avatar
LadyGeek
Site Admin
Posts: 71307
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Playing with Callan's periodic tables of investment returns

Post by LadyGeek »

Sold. We now have a draft page for review: User:LadyGeek/Callan periodic table of investment returns

I thought "coefficient of variation" should be part of the new section. I also added a sentence explaining that the "Agg Bond" asset class had the best risk vs. return tradeoff.

All wiki editors are welcome to edit the page. Readers can supply comments here.

Update: The draft page has been removed, see below.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

LadyGeek wrote: Fri Sep 07, 2018 3:21 pm Sold. We now have a draft page for review: User:LadyGeek/Callan periodic table of investment returns

I thought "coefficient of variation" should be part of the new section. I also added a sentence explaining that the "Agg Bond" asset class had the best risk vs. return tradeoff.
Looks great, nice work! Just one nitpick: personally, I'd remove the last two sentences of the CV explanation (the CV ranking, and the comment about the best tradeoff). This is extraneous text, it is obvious from the table and the explanations provided by the first two sentences. This would also avoid updating mistakes (the CV ranking sentence is wrong, it was probably not updated since the 2017 version). And we don't want to make too big of a deal about the CV metric here, the focus should be on the periodic table. Simple is (more) beautiful, and easier to maintain.

One last nitpick: I would reorder the overview bullet point, putting diversification first. This is THE point that the Callan folks emphasize in their PDF file (rightfully imho).
User avatar
LadyGeek
Site Admin
Posts: 71307
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Playing with Callan's periodic tables of investment returns

Post by LadyGeek »

Nits picked. The page is now "live", see: Callan periodic table of investment returns

Dispersion graph: I added a sentence to explain how to read the graph.

"Coefficient of Variation": I added a summary sentence, intentionally putting the lowest rank first - it's the most diversified asset class.

We can always make further changes.

=====================
OK, back to discussing your spreadsheet. :)
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
Topic Author
siamond
Posts: 5792
Joined: Mon May 28, 2012 5:50 am

Re: Playing with Callan's periodic tables of investment returns

Post by siamond »

LadyGeek wrote: Fri Sep 07, 2018 7:30 pm Nits picked. The page is now "live", see: Callan periodic table of investment returns.
I updated the Google Sheets for the dispersion graph and stats table, adding the missing HY Bonds, updating some out-of-sync numbers, focusing on the past 20 years (i.e. the most recent Callan's table), and adding more automation to ease future updates. And I did some minor rewording of the wiki page. I think we're good with it for now.
User avatar
LadyGeek
Site Admin
Posts: 71307
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: Playing with Callan's periodic tables of investment returns

Post by LadyGeek »

I have further revised the Google Sheets:
  • Dispersion graph: Fixed permissions so that anyone with a link can view the dispersion graph directly
  • Stats table:
    • Reordered the sheets so the "coefficient of variation" is seen first.
    • Added a date range to the title block
    • Spelled out "Coefficient of Variation"
    • Added max and min return columns which can be copy-n-pasted to the dispersion graph. (Google sheets does not allow direct links.)
Finally, I created a new section "Create your own periodic table" describing siamond's spreadsheet and pointing to this thread. The spreadsheet is very useful as an educational tool.

New investors are encouraged to look at the spreadsheet. If there are any questions, feel free to ask in this thread.

See: Callan periodic table of investment returns

My only concern is that new investors may not understand the use of High yield bonds in a portfolio, but that is an entirely different discussion. :wink:
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
Post Reply