Backtesting for the Optimum HBPP Allocations

General Discussion on the Permanent Portfolio Strategy

Moderator: Global Moderator

User avatar
MachineGhost
Executive Member
Executive Member
Posts: 10054
Joined: Sat Nov 12, 2011 9:31 am

Re: Backtesting for the Optimum HBPP Allocations

Post by MachineGhost » Fri Jan 09, 2015 10:50 am

robtkatz wrote: Is there a reference that explains this in detail?  I don't know much about it.
I would not recommend doing it since you're a neophyte, but heres a white paper about it:

http://papers.ssrn.com/sol3/papers.cfm? ... _id=962461
"All generous minds have a horror of what are commonly called 'Facts'. They are the brute beasts of the intellectual domain." -- Thomas Hobbes

Disclaimer: I am not a broker, dealer, investment advisor, physician, theologian or prophet.  I should not be considered as legally permitted to render such advice!
WrightO

Re: Backtesting for the Optimum HBPP Allocations

Post by WrightO » Wed Mar 04, 2015 8:19 pm

The mysterious peaktotrough.com calculator...........  ;D

For the period 01-01-2010 to 03-05-2015, I allocate 100% to S&P500 and use  "None" for rebalancing (the other options are the defaults). The output gives some returns under "Tbill" as well as under "S&P500" and adds them together to get the total return. The first entry in the "Tbill" column is 0, so the nonzero growth in that column is from 0 initial investment! There are other instances of the same sort of behavior. What am I missing here? Can I rely on the output from the calculator?
 
Thanks,

Klaus
mukramesh
Executive Member
Executive Member
Posts: 165
Joined: Fri Sep 12, 2014 3:27 pm

Re: Backtesting for the Optimum HBPP Allocations

Post by mukramesh » Mon Mar 09, 2015 2:32 pm

You probably have 'Reinvest Div/Int' set to NO. All dividends get put into Cash (T-Bills) and then get added to your total returns.
bedraggled
Executive Member
Executive Member
Posts: 705
Joined: Sat Sep 13, 2014 4:20 am

Re: Backtesting for the Optimum HBPP Allocations

Post by bedraggled » Fri May 29, 2015 6:17 am

When I try a 33/33/33/1 portfolio from 1-1-1975 to today with 35/15 bands, the S&P 500 and T Bond components match each year for 40 years.  I tried several times.  This problem may exist with the 4x25 but I need to wrap up for the morning

What is the problem, please.

Thanks
bedraggled
Executive Member
Executive Member
Posts: 705
Joined: Sat Sep 13, 2014 4:20 am

Re: Backtesting for the Optimum HBPP Allocations

Post by bedraggled » Sat Jun 13, 2015 8:09 pm

Follow up from my last inquiry: has anyone noticed the same numbers when running an HBPP peaktotrough 40 year analysis?

THanks
Justin
Junior Member
Junior Member
Posts: 11
Joined: Fri Jul 31, 2015 7:44 am

Re: Backtesting for the Optimum HBPP Allocations

Post by Justin » Fri Jul 31, 2015 8:12 am

Long time reader, first time poster. Too many people to thank for informative posts over the years, just hoping I can start to contribute a bit to pay back my debt to you all.

Today I was reading here:

http://www.ritholtz.com/blog/2015/07/th ... dilemma-2/

There is a section towards the back which reminded me of this fantastic thread where they say the global equity and debt market caps are respectively 48 and 41 trillion dollars. If you include gold as 9 trillion dollars worth, and then M3 of 18 trillion then you get the following ratios:

Equities: 41%
Bonds: 35%
Gold: 8%
Cash: 16%

Your mileage may vary but I thought this might be of interest.
User avatar
mathjak107
Executive Member
Executive Member
Posts: 4456
Joined: Fri Jun 19, 2015 2:54 am
Location: bayside queens ny
Contact:

Re: Backtesting for the Optimum HBPP Allocations

Post by mathjak107 » Fri Jul 31, 2015 8:16 am

although i am not keen on gold at all  i think it is a well balanced retirement portfolio  , in my opinion  i wouldn't go longer than intermediate term on the bonds though .
Last edited by mathjak107 on Fri Jul 31, 2015 8:18 am, edited 1 time in total.
AnotherSwede
Senior Member
Senior Member
Posts: 117
Joined: Mon May 11, 2015 10:24 pm

Re: Backtesting for the Optimum HBPP Allocations

Post by AnotherSwede » Fri Jul 31, 2015 8:56 am

Justin wrote: Equities: 41%
Bonds: 35%
Gold: 8%
Cash: 16%
Nice, a K�kkenbacher! I'll implement it after next stock market crash, and only if mortgage vs house value vs net worth is still in order. Until then - Desert portfolio!
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: Backtesting for the Optimum HBPP Allocations

Post by Kbg » Tue Sep 25, 2018 7:48 pm

User avatar
Tyler
Executive Member
Executive Member
Posts: 2066
Joined: Sat Nov 12, 2011 3:23 pm
Contact:

Re: Backtesting for the Optimum HBPP Allocations

Post by Tyler » Tue Sep 25, 2018 8:10 pm

Nice article. Thanks! It's always nice to see people talking about the variety of simple asset allocations available and sharing Harry Browne's ideas.

Now if we can just get these financial guys to stop using single arbitrary start dates to draw definitive conclusions about portfolios. I think a major draw of the PP is not the returns alone but how incredibly consistent it was no matter what timefreame you sample. They start to get there with the Sortino talk, but IMHO it goes deeper than that. The PP is simply an extremely dependable portfolio.
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: Backtesting for the Optimum HBPP Allocations

Post by Kbg » Wed Sep 26, 2018 6:58 am

Agree Tyler. Man I really wish there was a smartly leveraged cheap ETF of the PP. I’d be all in on it.
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: Backtesting for the Optimum HBPP Allocations

Post by Kbg » Wed Aug 14, 2019 6:54 pm

Thought I would post this. Conceptually one could implement the PP classically or with risk parity weighting. In the latter, cash is more something you dial in later to tame whatever mix you may be doing. (from 1978 using Portfolioviz)

CAGR Stdev Best Yr Worst Yr Max. DD SRatio
9.31% 9.36% 47.49% -12.38% -20.76% 0.53 (SBG eq wt - 33.33%)
9.48% 8.75% 36.51% -9.53% -16.75% 0.58 (32S/43B/25G)

If we throw 25% cash in and adjust proportionally

8.19% 7.04% 38.22% -5.61% -13.45% 0.53 (CSBG eq wt - 25%)
8.30% 6.59% 30.30% -3.55% -11.72% 0.58 (25C/24S/32B/19G)

Take the just above and merge the LTB and Cash into ITBs
8.45% 6.61% 31.43% -2.60% -11.57% 0.60

OR Take the top mix and sub ITBs for LTBs (32S/43ITB/25G)
8.80% 7.68% 39.71% -5.77% -15.08% 0.57

Looks like a risk parity PP is mo bettah in all cases...course it takes a little more work
User avatar
sophie
Executive Member
Executive Member
Posts: 1959
Joined: Mon Apr 23, 2012 7:15 pm

Re: Backtesting for the Optimum HBPP Allocations

Post by sophie » Thu Aug 15, 2019 7:26 am

Interesting question & reporting - thanks guys!

Applying what I know about statistical analysis to this question...I think you'd have run a permutation test, also known as a bootstrap. I would do this: Take each year's gain and average it with the year before and year after (since a given year's performance is not independent of adjacent years). Randomly shuffle each year since 1972 (or 1974 if you're afraid of the artificial gold spikes in 1972-73), then string together the first 15 years that come up to produce your outcome statistics. Do this several hundred times, or as long as it takes to get a stable distribution. Hope one of you has some time to do this ? (Sadly I don't.)

Going further than that, I'd label each year according to the economic condition represented i.e. prosperity, deflation, inflation, recession. Make sure each of these is represented in the random sample according to historical distributions.

You'd then have to run this for each combination of asset %'s to determine the optimal set. I believe this is what HB did at some point, which is how he arrived at the 25x4 percentages. I'm not sure if he balanced the economic conditions though, so he may have overweighted inflation & recession. That's why the Golden Butterfly is an interesting variation, to me: it recognizes that prosperity has dominated long enough that is not an unreasonable expectation that it will continue to do so going forward.
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: Backtesting for the Optimum HBPP Allocations

Post by Kbg » Thu Aug 15, 2019 10:23 am

For me I’ve long thought the actual secret sauce of the PP is the premise that we can not predict...ergo a lot of fancy statistical analysis flies in the face of the premise. We do have history and there we have to make an assumption that future will resemble past.

Good moment for full disclosure...to really properly have done the analysis for risk parity I would have ran the to date st dev each year and tweaked port%s slightly.

25% each is near zero predictive...but back to history, asset to Econ condition was based on historicals.

My backtest assumed the same assets PP came up with and that long term volatility was a tad more “neutral” given some assets have more jet fuel than others.
User avatar
vnatale
Executive Member
Executive Member
Posts: 9423
Joined: Fri Apr 12, 2019 8:56 pm
Location: Massachusetts
Contact:

Re: Backtesting for the Optimum HBPP Allocations

Post by vnatale » Mon Jan 27, 2020 12:10 pm

Tyler wrote:
Tue Sep 25, 2018 8:10 pm
Nice article. Thanks! It's always nice to see people talking about the variety of simple asset allocations available and sharing Harry Browne's ideas.

Now if we can just get these financial guys to stop using single arbitrary start dates to draw definitive conclusions about portfolios. I think a major draw of the PP is not the returns alone but how incredibly consistent it was no matter what timefreame you sample. They start to get there with the Sortino talk, but IMHO it goes deeper than that. The PP is simply an extremely dependable portfolio.
Tyler,

I know you are a person of great humility so it's not easy for you to praise yourself.

But am I correct to assume that all the work you have done at your web site subsequent to the start of this Topic now supersedes just about all the information in the prior posts here?

In other words, you've come up with better, more refined models which use better information?

That'd be my best guess!

But I'm also almost certain you'll downplay what you've accomplished.

Vinny
Above provided by: Vinny, who always says: "I only regret that I have but one lap to give to my cats." AND "I'm a more-is-more person."
User avatar
Tyler
Executive Member
Executive Member
Posts: 2066
Joined: Sat Nov 12, 2011 3:23 pm
Contact:

Re: Backtesting for the Optimum HBPP Allocations

Post by Tyler » Mon Jan 27, 2020 12:27 pm

vnatale wrote:
Mon Jan 27, 2020 12:10 pm

But am I correct to assume that all the work you have done at your web site subsequent to the start of this Topic now supersedes just about all the information in the prior posts here?

In other words, you've come up with better, more refined models which use better information?

That'd be my best guess!
There are lots of smart people on this forum, and I wouldn't claim that anything on PC automatically supercedes the opinions of anyone other than myself. But I've spent countless hours over the years improving the quality of my data and tools, and my analysis skills have definitely gotten more sophisticated over time. One of the perks of constantly reminding yourself to stay humble is that you never get so over-confident that you shut yourself off to new information. So I continue to learn new things over time and am happy to share.
Prakh
Junior Member
Junior Member
Posts: 1
Joined: Thu Aug 06, 2020 6:46 am

Re: Backtesting for the Optimum HBPP Allocations

Post by Prakh » Thu Aug 06, 2020 6:47 am

[spam content removed]
User avatar
Kriegsspiel
Executive Member
Executive Member
Posts: 4052
Joined: Sun Sep 16, 2012 5:28 pm

Re: Backtesting for the Optimum HBPP Allocations

Post by Kriegsspiel » Thu Aug 06, 2020 6:52 am

Prakh wrote:
Thu Aug 06, 2020 6:47 am
[spam content removed]
Ahhh, it was kind of nice with the new member's email thing being turned off.
You there, Ephialtes. May you live forever.
Post Reply