Portfolio Charts

[Greg]

So I added a new section to the site dedicated to the individual Assets themselves.  Take a look.  Of course this goes way beyond the PP assets, but perhaps people will find it helpful for their VP as well as their PP.  I could also use some extra sets of eyes on stuff like this, and I've found the group here to be particularly knowledgeable when it comes to the finer points of investing.  No need to clutter the thread with feedback, but if you spot a mistake or think I missed or flubbed something please PM me. 

You'll also probably find the working example in this post to be interesting from a PP perspective.  Sometimes it's hard to explain to people why correlations are important, but I took a decent shot using stocks and gold as an example.

Speaking of gold, I'd like to add a few notes on the top places to buy physical bullion.  What might be considered the top-3 most popular reputable dealers?

I've gone with Goldmart and APMEX in the past.

[ochotona]

Speaking of gold, I'd like to add a few notes on the top places to buy physical bullion.  What might be considered the top-3 most popular reputable dealers?

Texas Precious Metals, Shiner, TX

[Reub]

I've used Goldmart, but not recently. They were reliable but slow.

[Kriegsspiel]

I've used Goldmart and Colorado Gold.

[dragoncar]

Do you use daily data?  How hard would it be to add a 3x daily option and expense ratios?  Depending on how you do it, seems like it could be as easy as using the 1x data and multiplying by three on a daily basis.

[Tyler]

I only have annual numbers, so dailies are out.  I've thought about adding expense ratios, but am not too keen on constantly monitoring individual funds for updated accurate info.  So the links are a start, and maybe I'll work something out over time.

Thanks, all, for the gold dealer info.

[lordmetroid]

Another feature which I would think it would be cool is to have the calculations hyperlinkable.

Probably not possible using embedded excel sheet but if a Javascript or a server side calculation was used, it could be done linkable and the the URI specify which asset and which proportions.

[Tyler]

Yeah, that would probably require a much bigger change  than I'm willing to make right now, but I do really like the idea. 

[PP67]

Tyler, I was wondering if you had given any more thought to a suggestion I made a while back (below)?

One suggestion: Would it be difficult to post the last year few years of performance on the funnel chart (ie last 12 months under Year Invested 1, last 24 months under Year Invested 2 and so on sort of like you do on your Hurricane charts.  The would give an indication of how the portfolio has been performing recently relative to the full 42 year history.  I've been posting these values by hand (taken from PeaktoTrough) but it seems like you have those values adjusted for inflation.  For instance, per PeaktoTrough, the last 12 months CAGR is -1.32, the 2 yr CAGR is +3.62, the 3 yr CAGR is +1.18, 4 yr CAGR is +2.17 and 5 yr CAGR is +4.44 which if you post it on the funnel chart (a dot for each year) all fall within the range of the full life PP (albeit on the lower range at this time).  To me, it is an indication that the portfolio is still operating within the range one could expect.

Thanks again for all your incredible work!

[Tyler]

Short answer: yes.  I've been experimenting with it a bit.  Now that the two large January updates are out, I should have time to revisit.  Thanks for the reminder!

[lordmetroid]

Another portfolio for your site:
https://www.youtube.com/watch?v=8rTBEZSL7-4

[ochotona]

Or post recent year performances in very different colors

[Tyler]

After a bunch of experiments I'm not sure modifying the Funnel chart is the best way to show the data, but I'm working on a new calculator to address individual annual returns compared to history more directly.  The goal is to address the question "was last year normal" in a straightforward and helpful way.  I'll let you know when it's up. 

[lordmetroid]

After a bunch of experiments I'm not sure modifying the Funnel chart is the best way to show the data, but I'm working on a new calculator to address individual annual returns compared to history more directly.  The goal is to address the question "was last year normal" in a straightforward and helpful way.  I'll let you know when it's up.

I think the funnel is very nice but until I had the theory explained to me on a different blog, I didn't fully comprehend what the funnel tried to show. Perhaps a more comprehensive explanation would be beneficial.

[Tyler]

The would give an indication of how the portfolio has been performing recently relative to the full 42 year history. 

While I realize this isn't the specific implementation you requested, I've made a new calculator that I think addresses your core question.  Check it out.

I started a thread in the main forum to discuss the PP implications in more detail, but just wanted to mention here that I really appreciate the feedback and suggestion.

[stuper1]

Tyler,

Would there be any chance of adding another asset to your charts:  zero-coupon long term treasury bonds (e.g., EDV or ZROZ)?  You would probably have to do some kind of synthetic calculation to get older data.  I'm not certain on the details of such a calculation, but it may not be very difficult.  I find EDV to be a nice way to add some pep to long-term treasuries.

Thanks for considering.

[Tyler]

Following up on some earlier discussions, I now have a tool that will find the set of portfolios with the tamest drawdowns for a given minimum return. 

http://portfoliocharts.com/2016/03/07/t ... investors/

For kicks, type in a 5.8% return and look at the top option.  Look familiar? 

[D1984]

Following up on some earlier discussions, I now have a tool that will find the set of portfolios with the tamest drawdowns for a given minimum return. 

http://portfoliocharts.com/2016/03/07/t ... investors/

For kicks, type in a 5.8% return and look at the top option.  Look familiar? 

Hi Tyler,

I've noticed some of the higher returning portfolios that have 9% to 10.33% CAGRs (albeit with high max DDs and bad worst years) use Emerging Markets and International Small cap as components; what Emerging Markets annual return or monthly return data are you using; if you are using IFA data (which is what the Bogleheads SIimba spreadsheet uses) it is dead wrong for EM from 1976 to 1987 and I can post explaining why (and giving the correct numbers all the way back to 12/31/1975 straight from the actual S&P IFC emerging markets indices) if you wish. It will reduce the CAGR returns somewhat for Emerging Markets since (unlike as per the IFA data) the early to mid 80s royally sucked for emerging markets.

[Tyler]

Yes -- all data is from the Simba spreadsheet.  If you have better data and can explain why, I'm all ears!  Send me a PM.  I've also started communicating directly with the guys that update the Simba spreadsheet, and am happy to share your info and get their side as well.

FWIW, your note on EM is exactly why I included the option to exclude assets.  Some people have similar questions about gold or other assets.  Interestingly, there's not a single portfolio above a 7.8% CAGR that did not include some combination if emerging markets, international small, and small cap value.  Excluding those three will bring things down to earth and get to the heart of the results. 

[D1984]

Tyler,

Would there be any chance of adding another asset to your charts:  zero-coupon long term treasury bonds (e.g., EDV or ZROZ)?  You would probably have to do some kind of synthetic calculation to get older data.  I'm not certain on the details of such a calculation, but it may not be very difficult.  I find EDV to be a nice way to add some pep to long-term treasuries.

Thanks for considering.

I posted zero-coupon annual data back to 1986 or so (using the Bentham Funds...now American Century funds) and then EDV and ZROZ and PEDIX and VEDTX starting in 2007; I'll see if I can dig up that post later tonight.

Treasury zeros (Treasury STRIPS) only existed from early 1985 onwards although banks like Merill Lynch were creating Synthetic strips (LIONS, TIGRs, etc) as far back as 1981 or 1982. Also, Kamakura has a synthetic yield spline method for estimating zero coupon yields based on coupon bearing bond yields and has done so for 10-year and 20-year bonds back to the mid 1950s; if one had access to their data and methodology (which will cost $) and access to the CRSP 20 and 30-year daily yield data from 1961 onwards (which is rather costly; even more so than the Kamakura data), one could easily construct a synthetic zero coupon yield for a fund that had the same 25 or 26 year duration as EDV.

I do remember reading an abstract from Kamakura that said that unlike the coupon bearing yield curve, the synthetic 20-year zero coupon yield curve never actually inverted in the late 1970s/early 1980s interest rate spike; for every month studied it was either slightly hump shaped (the middle to long maturities had higher rates than the very long and short maturities) or actually non-inverted and thus upward sloping (the very long maturities had higher rates than the long ones which had higher rates than the middle and short ones, etc) which is what you would actually expect to see in a highly inflationary and high-interest rate environment when people expected the dollar to lose 13% or more in value each year and rates to never go back down; can you imagine a 30-year zero coupon bond yielding upwards of 21% (which is what you would get in late 1981 in a non-inverted yield curve environment)? The value of a continually rolled over--to keep it at a constant maturity and duration--portfolio of Treasury zeros would have been obliterated from the 1972 or 1976 temporary rate lows to the 1981 highs (although in partial compensation the 1982 rate fall as inflation was broken would've been one of their best years ever....upwards of 160% annual gains IIRC based on some synthetic backtesting I've done just using coupon bearing bond data).

[D1984]

Yes -- all data is from the Simba spreadsheet.  If you have better data and can explain why, I'm all ears!  Send me a PM.  I've also started communicating directly with the guys that update the Simba spreadsheet, and am happy to share your info and get their side as well.

FWIW, your note on EM is exactly why I included the option to exclude assets.  Some people have similar questions about gold or other assets.  Interestingly, there's not a single portfolio above a 7.8% CAGR that did not include some combination if emerging markets, international small, and small cap value.  Excluding those three will bring things down to earth and get to the heart of the results.

No problem; PM sent.