"Impermanent" PP for the skilled number-crunchers

[Kevin K.]

Thought I'd share this link to a. blog which has elicited a long thread on Bogleheads. The guy uses the PP assets only but rebalances them frequently according to formulas which are mostly beyond my pay grade. He posts performance on a weekly (or even more often) basis.

https://breakingthemarket.com/welcome/

[pmward]

Does he actually give a TLDR to his formula anywhere? Or is it really spread out in tiny puzzle pieces all over his blog? I love to crunch numbers, but I don't think I have the patience to go through all his blog posts just to try to piece together what he's doing.

[Vil]

Does he actually give a TLDR to his formula anywhere? Or is it really spread out in tiny puzzle pieces all over his blog?

I stopped reading after his/her statement found on the blog:
Index investing works by automatically “rebalancing” multiple assets, and by rebalancing more frequently, you can easily beat the market.

Anyway, nice try.

[Vil]

Does he actually give a TLDR to his formula anywhere? Or is it really spread out in tiny puzzle pieces all over his blog?

He does not. Actually, I was pretty quick on replying to that. His strategy was summed up by someone on BH forum:

1) Pick a mix of uncorrelated assets. Stocks, Bonds, Cash and Gold are a good selection.
2) Assign an expected return (arithmetic / simple / linear) for each. As per this thread, this could be a fixed value, such as 10%, 5.5%, 4.1% and 0.5%, or a suitably long "rolling window".
3) Measure their recent ("current") volatility (standard deviation). The trailing 30-50 days is generally used. Convert that to an annual SD.
4) Subtract the SD (actually the SD^2/2) from the (airthmetic) annual return - this gives the geometric (log / real) return.
5) Weight the assets according to their geometric returns.
6) Include correlation. When the correlation between two assets increases to 1, reduce the weight of the asset with the lower geometric return.
7) Leverage - compare each asset to the risk-free return. Increase leverage if the geometric return is greater than the risk-free return.


Disclosure: I am highly allergic to market master gurus sharing plethora of academic-level thoughts, and tables of impressive CAGR with less to none of the actual input data they've used and more of the internal strategy details. It might be only me, but I do not see the reason of telling the world how great his results are beside he is doing some sort of market research before starting his own fund

[sophie]

This looks like a fancy form of momentum...

I don't doubt he might be on to something here, but it also sounds like a pretty high maintenance routine with the potential for major tax implications with all that trading.

Also, I played a bit on the website and can't find any "how-to" for the entire set of 4 assets on there...obviously he doesn't want to spread his knowledge around too much. I sense that he's going to start charging a subscription fee and maybe an advisory service.

It might be fun to try this out as a VP play inside a tax-advantaged account, though.

[Vil]

I sense that he's going to start charging a subscription fee and maybe an advisory service.

And that's not necessarily bad, given the amount of time and effort he invested.

It might be fun to try this out as a VP play inside a tax-advantaged account, though.

You may find this worth, then : https://breakingthemarket.com/the-ultim ... -strategy/

P.S. Obviously, as opposed to what I stated I did not left his page immediately . I was still intrigued, while still not finding too much implementation details following the theory.

[pmward]

Does he actually give a TLDR to his formula anywhere? Or is it really spread out in tiny puzzle pieces all over his blog?

He does not. Actually, I was pretty quick on replying to that. His strategy was summed up by someone on BH forum:

1) Pick a mix of uncorrelated assets. Stocks, Bonds, Cash and Gold are a good selection.
2) Assign an expected return (arithmetic / simple / linear) for each. As per this thread, this could be a fixed value, such as 10%, 5.5%, 4.1% and 0.5%, or a suitably long "rolling window".
3) Measure their recent ("current") volatility (standard deviation). The trailing 30-50 days is generally used. Convert that to an annual SD.
4) Subtract the SD (actually the SD^2/2) from the (airthmetic) annual return - this gives the geometric (log / real) return.
5) Weight the assets according to their geometric returns.
6) Include correlation. When the correlation between two assets increases to 1, reduce the weight of the asset with the lower geometric return.
7) Leverage - compare each asset to the risk-free return. Increase leverage if the geometric return is greater than the risk-free return.


Disclosure: I am highly allergic to market master gurus sharing plethora of academic-level thoughts, and tables of impressive CAGR with less to none of the actual input data they've used and more of the internal strategy details. It might be only me, but I do not see the reason of telling the world how great his results are beside he is doing some sort of market research before starting his own fund

On the whole I don't think it sounds like a terrible strategy... I mean for my VP quant trend allocation there is a big reason I also chose to use volatility as the trigger for buy and sell. Volatility is a great indicator of what the near term return is going to look like. During periods of high and/or increasing volatility return is usually negative walking forward, and in periods of low and/or decreasing volatility returns are usually positive walking forward. I think where I draw the biggest question is his #2. Assigning an arbitrary assumption of return is not exactly accurate. I can see what he is trying to do though, where risk parity tries to simply rebalance often to keep the risk of all assets even while completely ignoring return, he is trying to take both risk and expected return into account and rebalance based upon both. But if his expected return is off, it does kind of defeat the purpose. Even using historical data for return is likely not going to be very accurate. I do think he really should give his actual formula so people can actually look at it, test it, and critique it. I may dig a bit further into his blog this weekend to see if I can gain a better understanding out of curiosity.

Of note though, is that while his returns and drawdowns look good, they also don't look any better than a lot of trend following systems out there, like the ones Ocho uses from Paul Novell, and those systems are much easier to implement. At least something worth keeping in mind.

[Vil]

they also don't look any better than a lot of trend following systems out there

Indeed. And with that frequent re-balancing... well he stated one has to use a real discount broker and to have an account > 300K USD to make the most of it (i.e. daily rebalancing).

like the ones Ocho uses from Paul Novell, and those systems are much easier to implement. At least something worth keeping in mind.

By the way which is the Paul Novell's true web page ? I have seen several in the past and got slightly confused, did not check that later. But that reminds me to ask you (or should I better ask ocho ) - I mean is there a place where the logic behind the SPY-COMP signals is explained ? Thanks.

[pmward]

they also don't look any better than a lot of trend following systems out there

Indeed. And with that frequent re-balancing... well he stated one has to use a real discount broker and to have an account > 300K USD to make the most of it (i.e. daily rebalancing).

like the ones Ocho uses from Paul Novell, and those systems are much easier to implement. At least something worth keeping in mind.

By the way which is the Paul Novell's true web page ? I have seen several in the past and got slightly confused, did not check that later. But that reminds me to ask you (or should I better ask ocho ) - I mean is there a place where the logic behind the SPY-COMP signals is explained ? Thanks.

You can see on this page me and Ocho having a discussion about it as I was digging into the system, and I gave the basic algorithm for the strategy. Apparently he uses a different moving average than the 200 day for his paid members, but you could test and find if there is one you like better, and the links on this page are to his website: viewtopic.php?f=10&t=8992&start=468

[sophie]

Isn't adding to the lagging asset the same basic result as frequent rebalancing without the tax implications? I mean, if you didn't have new money coming in and were doing this inside a 401k, there are no tax implications anyway.

Not in this guy's approach. In his "rebalances" the asset proportions change over time. Sometimes he moves to 100% in one of the assets, for example. Like I said, a very active approach that would take a lot of work to maintain - in violation of Harry Browne's Rule 1. A retired person with a love of figures might enjoy this though.

[Vil]

You can see on this page me and Ocho having a discussion about it as I was digging into the system, and I gave the basic algorithm for the strategy.

Thanks for the link, got the overall idea, but the point about the economic indicators I do not really understand. Actually, I share the common(?) belief that markets are somewhat following their own way of interpreting things going on; as Dalio currently stated "there is a real economy and there is a financial economy, which are intertwined but different. The real economy and the financial economy each has its own supply and demand dynamics".

[pmward]

You can see on this page me and Ocho having a discussion about it as I was digging into the system, and I gave the basic algorithm for the strategy.

Thanks for the link, got the overall idea, but the point about the economic indicators I do not really understand. Actually, I share the common(?) belief that markets are somewhat following their own way of interpreting things going on; as Dalio currently stated "there is a real economy and there is a financial economy, which are intertwined but different. The real economy and the financial economy each has its own supply and demand dynamics".

I replied to this in the other thread, as that's a better place for the discussion. I think Ocho will likely have some good opinions to share on this as well. viewtopic.php?f=10&t=8992&p=196578#p196578