Leveraged Permanent Portfolio Experiment

[Dragline]

More a matter of initial timing on that than anything else -- if I would have entered this two or three weeks earlier, the returns would be double or more.
I was looking for maximum volatility in the individual components -- SPHB probably would have been better, depending on the timing. Honestly, they have not been volatile enough to induce rebalancing at "classic" PP intervals like I thought they would. I am wondering if I should do it more often anyway -- like whenever a component goes +/- 5%. That would have worked pretty well for this particular period.
Guess that's why its an experiment!
As far as inflation is concerned, the return on the equity (capital) component is essentially infinite, because its 100% borrowed money (i.e., there is no capital). For comparison, I could have said my capital investment was $1 -- then the return would be 96,900% right now, which takes care of inflation.
Yes, I know there are always better alternatives in hindsight depending on what month you look at, but the real purpose of this is simply to see if I can beat the cost of the debt on a consistent basis and thereby create a money machine without employing any capital.

[dragoncar]

Are you still doing this? How's it going?

[dragoncar]

Bogleheads has an interesting related discussion (leveraging a balanced fund) where you choose an AA with a high sharpe ratio and lever up to get your desired risk/volatility.

http://www.bogleheads.org/forum/viewtop ... 0&t=143037

[Dragline]

dc, I am not still doing what was described above, but do have a partially leveraged basket of assets that includes some stocks, but mostly LT government bonds, reits, gold, and investments in Lending Club.

But I'd rather not disclose all the details. (Because I'm shy.)

[fips]

For what it's worth and because so many people on this forum seem to be inclined to follow the general notion of building a permanent portfolio, here's a backtest using the ETFs

SPY - equity market
TLT - long-term bonds
SHY - short-term bonds
GLD - gold

General assumptions:
- commission/trade: 5 USD
- variable slippage
- transaction price: next average of hi and low
- rebal frequency (if sell rules are met): every 4 weeks
- rebalance if portfolio weight goes < 20% or > 30%

Key stats:
- annualized return: 6.02%
- max drawdown: -16.22%
- annual turnover: 41.17%

Not great, but easy to understand and simple to follow.
I have built more promising models though.

If you like to see any other variation of this, let me know.

[MarginVariation]

What are you utilizing for the backtesting? Is it a paid commercial software?

[Dragline]

Here are some longer-term comparisons: http://www.crawlingroad.com/blog/2008/1 ... l-returns/

[zb2]

here is a useful comparison of return/risk of PP against other asset allocations (PP has relatively lower drawdowns compared to others over last 40 years) - http://mebfaber.com/2013/07/31/asset-al ... ategies-2/

[fips]

MarginVariation,

Yes, I am using a commercial software: Portfolio123

I would not advocate using it from the view of a common investor and an ERE-perspective, unless you know what you are doing and either have a "critical mass" of capital to invest for yourself or if you know how to market your ideas.

Best,
fips