PP Inspired Leveraged Portfolios

A place to talk about speculative investing ideas for the optional Variable Portfolio

Moderator: Global Moderator

hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Sat Jun 19, 2021 10:25 pm

StrategyDriven wrote:
Sat Jun 19, 2021 9:00 pm
hydromod wrote:
Fri Jun 18, 2021 6:34 pm
As a matter of curiousity, how do you model slippage from trades? I've noticed that slippage tends to ding returns noticeably with some of my more frequent rebalancing extravaganzas, especially with 1x funds.
I do not account for slippage. I do track ETF prices for full month and don't account for any rebalancing at the start of the new month simply out of convenience, this way I can just use the more easily available monthly total performance data. I personally do my trades at noon the last day of the month so I'm super close to actual reported data, but even trading the opening morning as AllocateSmartely did a study a while back, it has a negligible impact over time, flip a coin as to whether any specific month turns out better or not, tends to neutralize over time.
Thanks.

I end up doing some random selection of trade values between the open, high, low, and close for the day, which makes each run have a bit different CAGR. I worked out some ways to probabilistically extrapolate open, high, and low for mutual funds from the later ETFs, but haven't implemented them yet. Probably overkill; the bid-ask spread is the systematic slippage effect, but I have no idea how to find or estimate historical values.
User avatar
StrategyDriven
Senior Member
Senior Member
Posts: 135
Joined: Thu Jul 09, 2020 5:58 pm

Re: PP Inspired Leveraged Portfolios

Post by StrategyDriven » Sat Jun 19, 2021 10:27 pm

hydromod wrote:
Sat Jun 19, 2021 10:25 pm
StrategyDriven wrote:
Sat Jun 19, 2021 9:00 pm
hydromod wrote:
Fri Jun 18, 2021 6:34 pm
As a matter of curiousity, how do you model slippage from trades? I've noticed that slippage tends to ding returns noticeably with some of my more frequent rebalancing extravaganzas, especially with 1x funds.
I do not account for slippage. I do track ETF prices for full month and don't account for any rebalancing at the start of the new month simply out of convenience, this way I can just use the more easily available monthly total performance data. I personally do my trades at noon the last day of the month so I'm super close to actual reported data, but even trading the opening morning as AllocateSmartely did a study a while back, it has a negligible impact over time, flip a coin as to whether any specific month turns out better or not, tends to neutralize over time.
Thanks.

I end up doing some random selection of trade values between the open, high, low, and close for the day, which makes each run have a bit different CAGR. I worked out some ways to probabilistically extrapolate open, high, and low for mutual funds from the later ETFs, but haven't implemented them yet. Probably overkill; the bid-ask spread is the systematic slippage effect, but I have no idea how to find or estimate historical values.
It wouldn't be too difficult to do something like a spread difference, but I don't know that I think it's worth diving into.
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Sat Jun 19, 2021 11:22 pm

vincent_c wrote:
Sat Jun 19, 2021 8:10 pm
hydromod wrote:
Sat Jun 19, 2021 12:37 pm

Approach

With all that said, the strategy is really quite simple. I use a minimum variance optimizer for a portfolio of N assets, given a historical variance-covariance matrix for the assets, to develop the weights for the assets that would minimize the portfolio variance given the variance-covariance matrix. The optimizer is constrained by the requirement that the weights are between 0 and 1, and sum to 1. I don't provide expected returns.
Correct me if I'm wrong but your strategy, in short, is just a minimum volatility allocation for a specified group of assets.

You have to realize that when I asked for the simplest possible explanation of the strategy I was expecting a high level answer and really what you wrote is far from simple. Anyway I did read through but you will have to forgive me later if I missed something because the posts are quite long.

I'd like to ask about your chosen assets for the "N assets".

I am confused because you've mentioned the Russell 2000, QQQ, and TLT but you also mention assigning risk according to categories and you listed gold in there but it wasn't clear whether you've actually allocated to gold. Can you give me an idea of what your algorithm ends up with as asset allocation?

The example with the three assets was simply a reply to the rather different MAX PAIN approach that StrategyDriven pointed out. Both do very well over time, but they have very different volatility patterns.

Also, what are the reasons for diversifying between various equity ETFs?
vincent_c wrote:
Sat Jun 19, 2021 8:19 pm
hydromod wrote:
Fri Jun 18, 2021 1:10 pm
I'm currently in a UPRO/TQQQ/UTSL/DRN/TMF portfolio with 2/3 of risk to equities, but I'm considering running a URTY/TQQQ/DRN/TMF portfolio to (i) simplify, reduce correlations, and take advantage of barbelling, and (ii) assigning 3/4 of risk to equities to slightly boost returns.
I totally missed this from your original post because it was so long and I had no clue what UTSL and DRN until I looked it up. I also had to look up URTY. I would think even some people won't immediately recognize UPRO, TMF and TQQQ (although they could probably guess). Anyway It might have been easier if you gave the references as well especially in the first post.


Sorry - I hate when folks don't identify funds, and there I go doing it myself.

  • S&P 500: 3x = UPRO
  • NASDAQ: 3x = TQQQ, 1x = QQQ
  • Russell 2000: 3x = URTY, 1x = IWM
  • Long-term treasuries: 3x = TMF, 1x = TLT
  • Utilities: 3x = UTSL
  • Real estate (REITs): 3x = DRN



But even knowing this I am even more confused about the strategy but also can immediately think of some things to discuss as well. What is the theory behind why such an allocation or dynamic allocation will be an all weather approach (which is what I think of when I think about PP inspired approaches).
Sorry. I'm rarely mistaken for a terse writer, one of my failings. I'm still feeling out level of knowledge here - it's a wide spread on the boglehead forum.

The strategy is based on minimum variance, but layers a clustered risk budget on top of this. I started with inverse volatility, which basically assumes correlations are zero; that seems to work fairly well too, and the risk budget logic is identical.

The risk budget allows me to nudge how much assets contribute to the portfolio, while still honoring changes in volatility.

I've been testing various combinations of between 1 and 7 assets at once to understand algorithm behavior and test robustness. Typically various index funds and sector funds.

Generally I use two clusters: risk and safe. Risk accounts for equities; sometimes I drop REITS in too, or make them a separate cluster. Safe accounts for things like treasuries and bonds. I can test with an additional cluster for gold. I can partition between US and international, although I haven't taken that very far.

I find that using clusters makes it easy to automatically partition risk to the assets as I test different combinations.

In terms of overall allocation strategy, I think that equities are likely to be the primary driver of returns going forward, and I tend to use several equities at once. I try to select a suite that has low correlations but generally good returns. Even better if they have different economic cycles, so that they are out of phase.

I started with S&P 500 (UPRO) and long-term treasuries (TMF), because that was the entirety of the HFEA. Folks started talking about juicing returns with NASDAQ (TQQQ) and diversifying with other funds, which got me thinking about being more systematic.

Utilities (UTSL) were brought up because of the low correlation to UPRO and TQQQ. Some have considered utilities as a partial replacement for gold in the golden butterfly portfolio.

REITs (DRN) was brought up as a good diversifier with good returns as well.

Gold was brought up. This is a potentially contentious topic, but I generally haven't seen much benefit to including gold in a 3x portfolio since the 1980s (my backtesting period), given the lack of returns and the current lack of a 3x ETF. The 2x ETF takes up too much space to balance volatility, which cuts returns significantly.

I've tried a bunch of others, mostly 1x funds and 3x sector funds, that could be incorporated, but I couldn't find a justification for including them preferentially. Much of my testing looked at robustness to crashes and overall returns.

I think that the clustered risk budget minimum variance approach lends itself well to an all weather approach. One might select levered funds and weights that are more tailored to the all weather philosophy, boosting returns without necessarily raising portfolio volatility unacceptably. I would welcome discussion on that.
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Sun Jun 20, 2021 2:15 pm

vincent_c wrote:
Sun Jun 20, 2021 1:02 am
Thanks and welcome to the forum by the way.

I have a better idea of what we're talking about now and I think there are really two areas of discussion here.

The first is the general quantitative approach to begin with and Mark pointed out some of the key problems that are shared by many such approaches. What drew me to the PP in the first place was that it was an agnostic approach that isn't based on a backtest. You can come up with the PP allocation which is based on allocating to cover against any changes in the the economy that affects credit and duration risk and then backtest it to show you how it performed over any period which is what I think backtests should be used for.

You even mention that StrategyDriven had another strategy that does well over time and I think you'll find most strategies that are derived from backtesting will have this feature.

My feeling of what you're trying to do, is the opposite of the KISS approach so you're adding in complexity and perhaps trying to convince even yourself that it's a good thing to do whereas the main problem is that the complexities can sometimes hide what is a fundamentally flawed approach in the first place.

It's not only the math that is the problem though so the second thing is whether the chosen assets themselves are good.

One thing I like about the PP is that each component has very clear risk exposure and that the addition of each component increases the sharpe ratio of the portfolio. PP investors can actually seem hypocritical at times because I can tell you that I wouldn't invest 100% of my money into each component of the PP but I need to look at it as a whole.

However, when evaluating other strategies, you have to look at the individual components and consider whether they are good investments, not from a macro point of view, but just based on how the fund is set up and what it invests in. For example I would love to invest in "real estate" via a fund, but I just can't find one that actually invests in the things I want. In my opinion you can't take two bad investments, put them together to make a good investment.

The PP is different because each component can be invested in through the most liquid, cheapest, most efficient ways possible and the assets themselves represent the purest of risks in my opinion. For example, on the Bogleheads forum the total stock market and the S&P500 are viewed pretty much the same but in terms of liquidity the S&P500 has a lot more liquidity. ETFs often create a false sense of liquidity when you have to look at the underlying assets for true liquidity.

In other words, putting all your money to short or long gold may be a bad idea. But the concept of allocating to a store of value isn't necessarily a bad idea and the way to do it, whether physical, derivatives, or various ETFs and funds all have their merits and there are good choices and bad choices when it comes to implementation.

In summary, I do not think that any randomly selected group of funds where you test for their variance/co-variance using any kind of look back will come up with a system that can be extrapolated to the future. You have to start with an agnostic approach.
I suspect that we are a lot closer than it may have come across to you.

I don't at all disagree that it is an entirely appropriate approach to select the portfolio components first, based on your perception of the type of risk covered and how the fund is assembled/managed. Once the assets are decided upon, then the entire portfolio is partitioned among the selected assets.

Apparently it sounds as though I am suggesting approaching it by sticking together assets until I get good backtested combinations.

In actuality, what I have been trying to do is work through how to assemble a robust and reliable portfolio allocation approach regardless of the asset combinations, up to and including leveraged assets. I want to have confidence in how to manage the portfolio allocation, so I can focus on exactly which funds are included without worrying about how to manage them.

General approaches to portfolio allocations

It seems to me that there are basically three abstracted flavors for portfolio allocation: (i) decide on a fixed allocation, perhaps sliding a bit over time based on changing risk appetite (buy-and-hold with rebalancing); (ii) use market signals to swap assets in and out (momentum approaches); and (iii) maintain a set of assets in the portfolio, but adjust weights according to market conditions (e.g., flavors of risk parity). All three general approaches have their strengths and weaknesses, IMO. I've been particularly interested in how to handle 3x LETFs, just from the introduction in bogleheads.

It sounds like you are a proponent of the first approach. There's a bunch of good to the approach, especially in taxable accounts, and that's probably the best approach for those who want to remain hands off and those who have behavioral tendencies (e.g., fear, tinkering, etc.). However, somehow you need to decide on exactly what allocation to assign to each asset. It sounds like you favor either (i) selecting the allocation without recourse to backtests, and confirming that it would have performed acceptably (basically a one-time confidence-building test), or (ii) selecting the assets, and using repeated backtests to adjust the allocation to better meet some predetermined metrics. So essentially you are setting the asset allocation based on the risk profile over a long period of time (perhaps many decades).

In the second approach, you basically invest in assets for a fraction of time, rather than always investing the asset as a fraction of the portfolio. Assuming that the expected returns are constant in time, you should get similar returns if you have the same fraction of time and fraction of the portfolio. The devil is in the details of selecting the signal between assets; it seems hard to keep a good signal working for very long.

The third approach essentially combines the two by sliding the allocation weights around in time; some flavors will add and drop from a selected universe of assets, other will keep all assets active at all times. Again, the devil is in the details of how to slide around the allocation over time. In my proposed approach, I estimate the asset allocation for the next few weeks based on the observed risk profile over the previous few months. If something changes, the risk profile updates in a couple of weeks.

The decision on approach becomes exacerbated once leverage is introduced to juice returns, because the portfolio deviates from whatever allocation it's set to much faster than in a 1x portfolio. You can get very significant drift over a year in a 3x portfolio, which may very much change the portfolio risk profile.

In the first approach, you move up rebalancing frequency to monthly or quarterly or by bands. In the second approach, you reassess the allocation each month. In the third approach, you reassess the allocation periodically or perhaps by bands.

So as a practical matter, you have to take some action fairly frequently with 3x LETFs to maintain a risk profile. With a fixed allocation, that is simply pressing a rebalance button in M1 Finance. With the adaptive allocation, I first have to take the additional steps of (i) running a Matlab script that acquires the most recent fund histories and recalculates weights, and (ii) typing in the new allocation targets in M1. The benefit of these additional steps is a somewhat smoother ride (at least in every backtest I've run) and comparable or better returns, which gives a better Sharpe ratio.

It would not be worth it to me if the process weren't automated at least to this extent. I'd be even happier in some ways if I could fully automate it, but that could have all sorts of nasty side effects.

Comments on variance

I think that objecting to using variance in finance has a place. That place is in estimating the return frequency of extreme and rare events. Clearly extreme events cannot be properly characterized using variance.

But nobody tosses out variance entirely. Most metrics for portfolio performance use variance. Even something as mundane as a Sharpe ratio is theoretically useless (inversely proportional to the sqrt(variance)), yet you invoked it in your comment about improving portfolio performance without (I presume...) a second thought.

I think that it is actually quite reasonable to use variance as part of asset allocation strategies. After you've come to peace that you've selected the portfolio assets to adequately withstand tail events within the range of asset weights that you allow in the portfolio.

In practice, the extreme range of weights over 40 years for a portfolio with just S&P500 and LTT might swing between a 20/80 to an 80/20 portfolio, usually 40/60 to 70/30; how is this unreasonable? What would be different and how would it be better if tails were completely taken into account (given that you can't predict a priori when a tail will occur)?

Comments on asset selection

Asset selection is the point I'm fine-tuning now, because I'm pretty confident that the allocation approach itself is robust enough for my purposes and I can keep up with the updating.

To me the key attribute of all-weather type portfolios comes from the smooth ride, where maximum returns are sacrificed to maintain a steady return stream from the fraction of the portfolio that is suited to current economic environment.

Philosophically, to me a smooth ride with very good returns is preferable to trying for excellent returns with a more volatile portfolio. So I am conceptually aligned with the key attribute of all-weather type portfolios.

Given my personal constraints, in which the portfolio in question must be siloed and a small fraction of my overall portfolio, for now I'm perfectly at peace with keeping to an asset universe of 3x LETFs consisting of just equities and treasuries. Under that constraint, it makes sense to use a few low-correlation equities for diversification and long-term treasuries for safety. That type of portfolio tends to backtest well for at least the past 40 years, although clearly that's no guarantee for the future.

For the silo, I'm trying to reach this smooth ride goal with diversification and returns within the limited universe of 3x LETF assets. Given the limited nature of the portfolio, for the time being I'm willing to forsake other all-weather-preferred assets that might favor other economic conditions.

If I was doing this with my entire portfolio, I would probably do part with 3x and the remainder something similar with 1x (perhaps with a wider set of more conservative assets).

With all due respect, I think that the 3x part is a real game changer with respect to portfolio construction, as long as you can prevent huge losses. I believe I would be set during retirement if I can consistently maintain a steady stream of 5 percent returns from a 3x portfolio even during a doldrum, even with my limited set of assets that don't tune to all environments. That 5 percent is probably quite greedy if I'm pulling 15 percent long term.

I hope this wasn't too long again... It's a good back and forth I think.
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Sun Jun 20, 2021 10:50 pm

vincent_c wrote:
Sun Jun 20, 2021 8:29 pm
hydromod wrote:
Sun Jun 20, 2021 2:15 pm
what I have been trying to do is work through how to assemble a robust and reliable portfolio allocation approach regardless of the asset combinations
Regarding allocation, you are correct that I default to a fixed allocation. Because of this, the way I see it is you are either timing the market (actively or passively) or you are not. Rebalancing is usually a market timing event. I've given this some thought before and my view is that you have to constantly rebalance in order to avoid market timing and any longer frequency is just a proxy to constant rebalancing.

Any quantitative approach to rebalancing a fixed allocation is just choosing a market timing approach beforehand and any deviation from the allocation is riding momentum and whenever the rebalancing occurs that becomes market timing. What makes your approach interesting is that you are designing a dynamic allocation so I need some time to figure out whether this is a legitimate alternative to avoid the market timing rebalancing dilemma.

If it helps, I'm leaning towards simply rebalancing every 2 to 3 weeks. I'm pretty sure it wouldn't make much difference doing a bands approach, based on back testing and logic, but it's easier for me to just sit down on a Saturday and set up the next update. If I was holding a fixed allocation, I'd just rebalance on a recurring Wednesday afternoon every 2 to 4 weeks.

The reason isn't because all economic conditions have an equal probability outcome, the reason is because market reaction to economic conditions are unpredictable and at the smallest timescale where constant rebalancing may occur my preference is to assume an equal probability of a shift toward any particular market reaction to a shift in economic condition.

Now this is where I am curious. While this sounds good, I'm not exactly sure what this actually means in practical terms. If I were to do an experiment to test such a concept, what would I measure to calculate a shift in market reaction? A shift in economic condition?

Now I'm imagining a vessel with gas atoms buzzing about and bouncing off of each other, with the random impacts resulting in Brownian motion. I think that's an analog that some use for financial markets.

hydromod wrote:
Sun Jun 20, 2021 2:15 pm
t sounds like you favor either (i) selecting the allocation without recourse to backtests, and confirming that it would have performed acceptably (basically a one-time confidence-building test), or (ii) selecting the assets, and using repeated backtests to adjust the allocation to better meet some predetermined metrics. So essentially you are setting the asset allocation based on the risk profile over a long period of time (perhaps many decades).
Regarding backtests, all I meant was that after you choose a portfolio and determine that over a certain period of time you expect it to have certain risk and return profiles you could run backtests over any number of periods to see if in the past such a strategy would work and since you did not construct a portfolio using backtests then if the backtest revealed something you could go and take a look to see if some of your assumptions over that period were not correct.

I wouldn't tweak allocation based on backtest results except perhaps the amount of leverage that I require. I go by you need to have the need and willingness to take risk, so if you have a certain need then you need to assess whether you are willing to bear the risk. Sometimes you are limited by your willingness and you need to make up the shortfall by other forms of income.

The way I have chosen to determine allocation is to figure out my need and then my willingness. If this calls for leverage then a backtest could help determine that and I would also consider using the kelly criterion. To me leverage only adjusts the amount of cash or borrowing and does not affect the allocation of the assets. The actual allocation of the assets I agree you can have some quantitative approach to determine if you so wish, but the PP has its simplicity of allowing the investor to easily choose 20/30 or 15/35 rebalancing bands and 15/35 makes it very rare that a single asset would have to be rebalanced twice consecutively.

Need is such a difficult variable to predict. It ratchets up so easily, and life comes at you fast sometimes. I'm sort of working at getting to a baseline with some cushion, using part of the portfolio to preserve the baseline, then buttressing it with the rest of the portfolio to cover increasingly unlikely events and events.
hydromod wrote:
Sun Jun 20, 2021 2:15 pm
yet you invoked it in your comment about improving portfolio performance without (I presume...) a second thought.
Regarding sharpe ratio, you can theoretically improve sharpe ratio without using backtests. A sharpe ratio will increase if you combine an asset with a positive return that has negative correlation to your current combination of assets. A backtest can confirm whether the sharpe ratio was as you expected relative to other allocations in the past.

I was tweaking you about the Sharpe ratio. You divide by standard deviation, which is the square root of variance, which in theory can't be calculated.
hydromod wrote:
Sun Jun 20, 2021 2:15 pm
In practice, the extreme range of weights over 40 years for a portfolio with just S&P500 and LTT might swing between a 20/80 to an 80/20 portfolio, usually 40/60 to 70/30; how is this unreasonable?
Of course we both agree that you should not never rebalance for the reason you stated. What makes a variance based approach unreasonable is how that variance is determined. If it is selected based on cherry picking data then it's meaningless. If it is based on historical data without considering why such a variance existed then it is meaningless. There has to be both historical variance and a fundamental cause for that variance and for such assumption to hold in the future for an investor to have confidence the variance will persist.

I guess I'm happy with an approximation using the last few months returns as a good enough estimate of the next few weeks for estimating allocations, because it can't give me a wholly unreasonable allocation and I'll update again soon. But yeah, you want to be increasingly careful with longer durations.
hydromod wrote:
Sun Jun 20, 2021 2:15 pm
Philosophically, to me a smooth ride with very good returns is preferable to trying for excellent returns with a more volatile portfolio. So I am conceptually aligned with the key attribute of all-weather type portfolios.
Given such an attitude toward investing, the simplest PP portfolio is not a bad choice and any leveraged portfolio or a quantitative approach may challenge your resolve when testing the limits of the strategy. Again we are back to tail risk here but it is something that is likely to occur over an investing lifetime.

Ah, the definitions for very good returns and smooth ride! I'm thinking of excellent returns as more than 30%/year, very good returns as more than 20%/year. How 3x changes one's perspective...

2020 was good that way, a bit of a love tap to alert me to my behavioral tendencies. I really had only started paying any attention to the market in 2019, just autoinvesting and scanning the quarterly 403b investment reports. I had a plan for using unemployment data as a macroeconomic indicator layered onto volatility, it worked really well back to the 1950s. Of course COVID was entirely unsuited and I tried ad libbing without a plan. Bad idea, but I got a small taste that I needed a framework to take decisions out of my hands.

I came up with a complement to drawdown that seems to suit my psychology, and may be especially useful with 3x. The drawback represents the amount of returns that a drawdown has consumed, measured in time since same portfolio value was first seen. So if a crash came after a bubble or spike, a big drawdown may still only represent part of a year that has been given back.

The image plots the drawback for an example portfolio with 3x small caps/nasdaq/REIT/long treasuries. The drawdowns suggest that I should expect a 25% drawdown every five years or less, and at least one or two 50% drawdowns in the next decade or two.

The drawbacks suggest that I'll quite frequently give back six or eight months, even a year or so. Drawbacks of 3 to 5 years should be expected with the big drawdowns. But it will be unusual for recovery to take more than a year or two.

I did this to remind myself that a drawback of 4 years may seem like a lot of money, but it may not be such a big deal when you still have the fruits of consistent returns of 15 to 25% up to 4 years previous and you only really need returns of 4 to 6%. Silver linings and such.

Image


hydromod wrote:
Sun Jun 20, 2021 2:15 pm
With all due respect, I think that the 3x part is a real game changer with respect to portfolio construction
I implement the PP using futures and I am leveraged so really you are right we are not far off. I don't understand why some people will use leverage but then buy SHV or SHY for example. I also don't understand why people will use 3x ETFs with high fees when you can get leverage from futures at the cheapest rate. Mark mentioned taxes in the US and that could be one reason however one should bear in mind not to let the tax tail wag the investment dog.

You were wondering if there was a 3x leveraged gold ETF. You could easily get 3x leverage using gold futures or options. You could also leverage more on stocks and bonds and just buy 3x the amount of physical gold. There are lots of things that you can do when you choose the PP assets to invest in.

For me, I'm comfortable with 3x LETFs because they are so easy to implement and I have no idea how to do futures or options. Basically just type in the desired allocation and punch the button to reallocate. I'm likely to be pretty happy with a consistent 10 to 15% returns. If I'm getting an average of 20 to 30% return a year, instead of 8 to 10%, I'm more than willing to give up 1% extra ER for the privilege. More power to those that can get the same returns and shave off the extra ER.
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Mon Jun 21, 2021 8:45 am

vincent_c wrote:
Sun Jun 20, 2021 11:06 pm
Can you convince me that there is only a 1% tracking error on the 3x ETFs?
Probably not.

Siamond and others on bogleheads have looked at tracking error. I think they pretty much concluded that ProShares is much better than Direxion in that regard, and I seem to remember that the more liquid funds are better. I seem to remember something about equities matching better than bonds. In general it seems like the first year or two has worse tracking issues, then things settle down.

I'd have to try and track down some of the posts to get estimates. I really don't remember where the decimal place is.

I remember concluding that LETFs are their own thing that is pretty close to tracking some market index but not quite. I just leave it at that, given how many different indices that exist nowadays.

A big question in my mind is in developing synthetic 3x ETFs, in particular incorporating borrowing costs. The recent record with 3x LETFs has such low interest rates, it hasn't been too much of an issue, but there are hints that there would have been more slippage and hidden costs at earlier times. So I take the synthetic LETFs prior to 2010 or so as being potentially pretty optimistic on returns.

I suspect that the funds run opposing bear/bull ETFs partly to even out borrowing costs. The details likely depend on assets to the + and - funds. So you may get anomalous tracking from that source. You'd need AUM histories to try and sort that out.

It's interesting to try a daily-rebalanced portfolio that is half +3x and half -3x for the same index. You'd expect these to monotonically decay because of the ER, but that hasn't always been the case, especially with the synthetic ETFs. If I remember correctly, there are times when you might have gotten positive returns; the bear fund gets a borrowing bonus that outweighs the borrowing cost for the bull fund.

Nowadays these borrowing effects are small compared to the daily fluctuations, of course, but may still affect long-term returns.
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Mon Jun 21, 2021 1:22 pm

vincent_c wrote:
Mon Jun 21, 2021 8:52 am
Does this not imply a 7-8% tracking error?
You're mostly showing the difference between tracking on a monthly basis and a daily basis, with the ER tossed in. When you aggregate the daily returns over a month, there's things like volatility decay that creep in. You can have perfect daily tracking but slippage on the monthly tracking.

UPRO tracks 3x daily movements. That's pretty consistent across LETFs.

I think I saw one or two funds that try to replicate monthly S&P behavior, but I misremember whether they were leveraged. I just remember that it just struck me as a potentially tricky fund to manage. It may have some utility for folks though.

I'm not fully understanding how futures and options work, but my impression is that they lean to the monthly returns. Some of the bogleheads tout that as an advantage for avoiding volatility decay.

Using PV in this way gives a quick and dirty estimate for 3x LETF returns, but for precision testing you should develop your own series and upload. Unfortunately this is no longer free; it was great while it lasted.
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Mon Jun 21, 2021 8:47 pm

vincent_c wrote:
Mon Jun 21, 2021 8:18 pm
Do you know that you are getting proper market rates on borrowing cost on the 3x ETFs without markup? Do you know anyone who has looked into it?
I believe that some of the bogleheads threads have discussed some of this. Certainly folks have tracked down the source of borrowing for some of the different LETFs.

There are a few threads in particular, the first is probably better for answering your question and the last is where I started getting ideas. There's a boglehead wiki on LETFs but it doesn't get to that detail.


One of the entries in the HEDGEFUNDIE thread here specifically compares synthetic monthly returns for the 3x S&P and 3x LTT; he found a 2% discrepancy for the 3x S&P of 2% over 10 years and an >8% discrepancy for the 3x LTL over the same period (mostly from the first 2 years). This kinda gets at one of your other questions.

I think Siamond would be the particular member that would be closest to your question, or would know who to ask. He's the one who maintains some daily, monthly, and annual backtesting spreadsheets and he is very concerned about such questions. He's very helpful in answering questions too.
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Mon Jun 21, 2021 8:54 pm

vincent_c wrote:
Mon Jun 21, 2021 8:18 pm
Also, if a fund is always increasing leverage when it’s gone up and decreasing leverage when it’s gone down then isn’t it always buying high and selling low? Over any given month daily prices are bound to have random fluctuations. How can this be a good thing?
LETFs reset the leverage each day.

My impression is that they hold the underlying securities, then work out swaps for the remainder of the leverage so that they can reproduce x times the daily movement. I think that they probably set the borrow interest rate with contracts covering several months. Something like the LIBOR overnight rate or such.
User avatar
Mark Leavy
Executive Member
Executive Member
Posts: 1950
Joined: Thu Mar 01, 2012 10:20 pm
Location: US Citizen, Permanent Traveler

Re: PP Inspired Leveraged Portfolios

Post by Mark Leavy » Mon Jun 21, 2021 9:17 pm

vincent_c wrote:
Mon Jun 21, 2021 8:18 pm
Do you know that you are getting proper market rates on borrowing cost on the 3x ETFs without markup? Do you know anyone who has looked into it?

Also, if a fund is always increasing leverage when it’s gone up and decreasing leverage when it’s gone down then isn’t it always buying high and selling low? Over any given month daily prices are bound to have random fluctuations. How can this be a good thing?
I sincerely admire your gentle, Socratic method. I don't have the patience.
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Tue Jun 22, 2021 12:24 am

vincent_c wrote:
Mon Jun 21, 2021 10:41 pm
I think we can take a step back from implementation methods and discuss the reasons for your input assets. You came to the PP forum, are you thinking of sticking the PP in?

I mean, I would love for you to test my chosen assets (the PP + usdcad exposure) and figure out using your allocation algo what the performance would have been and whether there has been any improvement in risk/return in the past.

Compare it with 50% usdcad exposure in combination with the 4x25 PP.
Honestly, I came in by happenstance mainly to say hi because I somehow saw some link to where folks here have been discussing 3x LETFs forever and I got a bit excited. A boglehead thread related to 3x risk parity got me started looking at finance and portfolio construction approaches in some detail, perhaps because I've been doing numerical modeling and risk assessment stuff for 3 or 4 decades and that was a hook I could start with. But with that said, I've only thought about finance for a year or two, I'm still in the intense learning stage, and I'm sure that I know very little overall.

Right now I'm trying to take in various ideas and portfolio construction approaches, and shake them to see if they make sense and see how I can use them effectively. So I'm eager to test various ideas, but at this point, I'll admit that I don't have a deep understanding of the PP philosophy. My main insight is from here.

If you specify the particular assets, I'd be happy to run them. If you have a backtesting dataset, I'd be happy using that for consistency, or I could try to cobble together a set. I'm presuming you want to go back a long time, I don't have a currency fund in my library, and I can only do gold and LTT to 1986 or so. Best data would be daily, but with a little work I could get by with monthly if I had an estimate for volatility during the month.

I'm suspecting that a risk parity/minimum variance type of approach may need some slight tuning to do a PP-based portfolio justice, though. The issue I run into running with cash or other low-volatility assets is that equal risk budgets tend to make the low-volatility assets dominate the portfolio. So the risk budget should be roughly inversely proportional to the volatility to get long-term averages to equal out.
D1984
Executive Member
Executive Member
Posts: 730
Joined: Tue Aug 16, 2011 7:23 pm

Re: PP Inspired Leveraged Portfolios

Post by D1984 » Tue Jun 22, 2021 1:37 am

hydromod wrote:
Tue Jun 22, 2021 12:24 am
vincent_c wrote:
Mon Jun 21, 2021 10:41 pm
I think we can take a step back from implementation methods and discuss the reasons for your input assets. You came to the PP forum, are you thinking of sticking the PP in?

I mean, I would love for you to test my chosen assets (the PP + usdcad exposure) and figure out using your allocation algo what the performance would have been and whether there has been any improvement in risk/return in the past.

Compare it with 50% usdcad exposure in combination with the 4x25 PP.
Honestly, I came in by happenstance mainly to say hi because I somehow saw some link to where folks here have been discussing 3x LETFs forever and I got a bit excited. A boglehead thread related to 3x risk parity got me started looking at finance and portfolio construction approaches in some detail, perhaps because I've been doing numerical modeling and risk assessment stuff for 3 or 4 decades and that was a hook I could start with. But with that said, I've only thought about finance for a year or two, I'm still in the intense learning stage, and I'm sure that I know very little overall.

Right now I'm trying to take in various ideas and portfolio construction approaches, and shake them to see if they make sense and see how I can use them effectively. So I'm eager to test various ideas, but at this point, I'll admit that I don't have a deep understanding of the PP philosophy. My main insight is from here.

If you specify the particular assets, I'd be happy to run them. If you have a backtesting dataset, I'd be happy using that for consistency, or I could try to cobble together a set. I'm presuming you want to go back a long time, I don't have a currency fund in my library, and I can only do gold and LTT to 1986 or so. Best data would be daily, but with a little work I could get by with monthly if I had an estimate for volatility during the month.

I'm suspecting that a risk parity/minimum variance type of approach may need some slight tuning to do a PP-based portfolio justice, though. The issue I run into running with cash or other low-volatility assets is that equal risk budgets tend to make the low-volatility assets dominate the portfolio. So the risk budget should be roughly inversely proportional to the volatility to get long-term averages to equal out.
Hi Hydromod (and Vincent_C),

I can get you data for the plain "USD index" (i.e. just the price of USD vs a basket of other developed country currencies) back to 1-1-1967 monthly and from 1-1-1971 daily if you'd like it. Note that to construct a simulation of the UUP ETF you would (IIRC...either of y'all or anyone else can please feel free to correct me if this is wrong) basically just use the monthly returns of the US T-Bill (taken from FRED) plus the monthly returns of the USD index; you would add them together rather than averaging them because what you are doing is simulating a futures position in the USD index backed by collateral (in this case, highly liquid and interest-rate risk free T-Bills). One you have this you could presumably use Canadian Treasury Bill Yields (I have these back to 1950 on a monthly basis if you need them) and the USD/CAD index to try and simulate the ETF FXC (I presume it would hold Canadian Treasury bills rather than just holding non-interest CAD sitting in a bank account or liquid reserve account with the Bank of Canada); having done this, you could perhaps do the 14-day free trial on PortfolioVisualizer using a throwaway email address and see how a yearly, monthly, or quarterly rebalanced long UUP/short FXC would perform.......unless I am understanding something wrong here and Vincent_C's long USD/short CAD portfolio was merely created by going long VFISX (or SHY) and short FXC?
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Tue Jun 22, 2021 8:12 am

D1984 wrote:
Tue Jun 22, 2021 1:37 am
vincent_c wrote:
Mon Jun 21, 2021 10:41 pm
I think we can take a step back from implementation methods and discuss the reasons for your input assets. You came to the PP forum, are you thinking of sticking the PP in?

I mean, I would love for you to test my chosen assets (the PP + usdcad exposure) and figure out using your allocation algo what the performance would have been and whether there has been any improvement in risk/return in the past.

Compare it with 50% usdcad exposure in combination with the 4x25 PP.
Hi Hydromod (and Vincent_C),

I can get you data for the plain "USD index" (i.e. just the price of USD vs a basket of other developed country currencies) back to 1-1-1967 monthly and from 1-1-1971 daily if you'd like it. Note that to construct a simulation of the UUP ETF you would (IIRC...either of y'all or anyone else can please feel free to correct me if this is wrong) basically just use the monthly returns of the US T-Bill (taken from FRED) plus the monthly returns of the USD index; you would add them together rather than averaging them because what you are doing is simulating a futures position in the USD index backed by collateral (in this case, highly liquid and interest-rate risk free T-Bills). One you have this you could presumably use Canadian Treasury Bill Yields (I have these back to 1950 on a monthly basis if you need them) and the USD/CAD index to try and simulate the ETF FXC (I presume it would hold Canadian Treasury bills rather than just holding non-interest CAD sitting in a bank account or liquid reserve account with the Bank of Canada); having done this, you could perhaps do the 14-day free trial on PortfolioVisualizer using a throwaway email address and see how a yearly, monthly, or quarterly rebalanced long UUP/short FXC would perform.......unless I am understanding something wrong here and Vincent_C's long USD/short CAD portfolio was merely created by going long VFISX (or SHY) and short FXC?
That would be great!

This entry is pointing out that it would be helpful for me to have a reference so that I properly understand how a long/short portfolio is implemented, especially in the way that vincent_c is intending.

D1984, I saw in your discussions with StrategyDriven that you had some extensive data going back years before my little stash and was hoping for some excuse to ask about it. Not necessarily at the moment, but perhaps in a while, if you are open. If not, no worries.
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: PP Inspired Leveraged Portfolios

Post by Kbg » Tue Jun 22, 2021 9:18 am

Can I humbly as the originator of this thread request we not replicate the Hedgefundie Bogleheads thread here? The level of discussion and detail in that thread is truly epic and does not need to be repeated in this forum. I have a hard time conceiving a point that hasn't been covered there in detail and often with data. Unfortunately it's so long now that its gold nuggets are lost in a pile of useless comments and minutiae. I think I've read every post, but my one critique is that some serious forest got lost in a continent's worth of trees. The basic principles of a leveraged portfolio are simple.

Don't add excessive risk...beyond 2x leverage additional performance benefits gets sketchy at best

3x ETFs and futures are NOT functionally equivalent. Each option has pros and cons
- Daily reset works extremely well for good trends either way if one does not rebalance frequently...better than futures
- Futures are cheaper, no question about it
- Path dependency is "THE" most important factor as to which option is better from a performance perspective...no one knows the future, therefore you only know the answer to "best" in hindsight
- Tax effects are probably more important than any of the above three
- Unless one has a very large account, futures are significantly more difficult to manage...with the advent of micro interest rate futures coming in the not too distant future there will be a legit small investor option of going the futures route (assumes using the PP's assets)

Correlation is critical and stocks/bonds get highly correlated during periods of rising interest rates...no one knows the future, therefore the "best" mix is known only in hindsight. (Commentary: I was stunned in the HF thread that what was clearly in the data was by and large blown off with a smug; "No, worries I can handle 25 years of underperformance no problem, I'm young." (And stupid))

My #1 recommendation is figure out your personal risk profile, study correlations and then dial in an asset mix you think you will be comfortable with over the long haul. Realize you are hoping history provides some guide to the future.

My mix:

70% BND
25% TQQQ
5% VIXY/VXX

I'm on year 7 of doing this. One can see over time my mix has evolved if they review this and the 2x leverage thread. Next time we get a good market tank I'll dial TQQQ up to 33% and BND down. Leverage is ~1.5x with a constant hedge.
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Tue Jun 22, 2021 12:47 pm

Kbg wrote:
Tue Jun 22, 2021 9:18 am
Can I humbly as the originator of this thread request we not replicate the Hedgefundie Bogleheads thread here? The level of discussion and detail in that thread is truly epic and does not need to be repeated in this forum. I have a hard time conceiving a point that hasn't been covered there in detail and often with data. Unfortunately it's so long now that its gold nuggets are lost in a pile of useless comments and minutiae. I think I've read every post, but my one critique is that some serious forest got lost in a continent's worth of trees. The basic principles of a leveraged portfolio are simple.

I feel it's incumbent on me to reply as the Hedgefundie instigator, especially since I seem to have barged in, don't know your forum dynamics, and have no desire to sidetrack your thread.

I think this is excellent advice. I ended up with a sidebar thread there just to document my evolving understandings in one place because of all the chaff. No need to reinvent anything here at all.


Don't add excessive risk...beyond 2x leverage additional performance benefits gets sketchy at best

No question about not adding excessive risk. I don't think that it's controversial that 2x has been shown to be a reasonable risk limit for individual funds in isolation. Respectfully, I'd perhaps quibble with this conclusion with respect to carefully selected combinations of funds, although I'm not advocating doing so without very good reason and careful thought.

3x ETFs and futures are NOT functionally equivalent. Each option has pros and cons
- Daily reset works extremely well for good trends either way if one does not rebalance frequently...better than futures
- Futures are cheaper, no question about it
- Path dependency is "THE" most important factor as to which option is better from a performance perspective...no one knows the future, therefore you only know the answer to "best" in hindsight
- Tax effects are probably more important than any of the above three
- Unless one has a very large account, futures are significantly more difficult to manage...with the advent of micro interest rate futures coming in the not too distant future there will be a legit small investor option of going the futures route (assumes using the PP's assets)

Correlation is critical and stocks/bonds get highly correlated during periods of rising interest rates...no one knows the future, therefore the "best" mix is known only in hindsight. (Commentary: I was stunned in the HF thread that what was clearly in the data was by and large blown off with a smug; "No, worries I can handle 25 years of underperformance no problem, I'm young." (And stupid))

My #1 recommendation is figure out your personal risk profile, study correlations and then dial in an asset mix you think you will be comfortable with over the long haul. Realize you are hoping history provides some guide to the future.

Agreeing with your recommendation without trying to sidetrack, this is a part that interests me, especially the correlations part. Could I ask for a bit of a pointer or reference or thread link as to precisely what you are referring to by correlations so I could take appropriate advantage of this advice?

And as a point of curiosity with respect to forum dynamics, some of the recent discussion in this thread seems to imply that using statistical approaches with finance is controversial or severely circumscribed among forum members, which would make using correlations controversial as well. Am I misunderstanding?


My mix:

70% BND
25% TQQQ
5% VIXY/VXX

I'm on year 7 of doing this. One can see over time my mix has evolved if they review this and the 2x leverage thread. Next time we get a good market tank I'll dial TQQQ up to 33% and BND down. Leverage is ~1.5x with a constant hedge.
D1984
Executive Member
Executive Member
Posts: 730
Joined: Tue Aug 16, 2011 7:23 pm

Re: PP Inspired Leveraged Portfolios

Post by D1984 » Tue Jun 22, 2021 11:45 pm

hydromod wrote:
Tue Jun 22, 2021 8:12 am
D1984 wrote:
Tue Jun 22, 2021 1:37 am
vincent_c wrote:
Mon Jun 21, 2021 10:41 pm
I think we can take a step back from implementation methods and discuss the reasons for your input assets. You came to the PP forum, are you thinking of sticking the PP in?

I mean, I would love for you to test my chosen assets (the PP + usdcad exposure) and figure out using your allocation algo what the performance would have been and whether there has been any improvement in risk/return in the past.

Compare it with 50% usdcad exposure in combination with the 4x25 PP.
Hi Hydromod (and Vincent_C),

I can get you data for the plain "USD index" (i.e. just the price of USD vs a basket of other developed country currencies) back to 1-1-1967 monthly and from 1-1-1971 daily if you'd like it. Note that to construct a simulation of the UUP ETF you would (IIRC...either of y'all or anyone else can please feel free to correct me if this is wrong) basically just use the monthly returns of the US T-Bill (taken from FRED) plus the monthly returns of the USD index; you would add them together rather than averaging them because what you are doing is simulating a futures position in the USD index backed by collateral (in this case, highly liquid and interest-rate risk free T-Bills). One you have this you could presumably use Canadian Treasury Bill Yields (I have these back to 1950 on a monthly basis if you need them) and the USD/CAD index to try and simulate the ETF FXC (I presume it would hold Canadian Treasury bills rather than just holding non-interest CAD sitting in a bank account or liquid reserve account with the Bank of Canada); having done this, you could perhaps do the 14-day free trial on PortfolioVisualizer using a throwaway email address and see how a yearly, monthly, or quarterly rebalanced long UUP/short FXC would perform.......unless I am understanding something wrong here and Vincent_C's long USD/short CAD portfolio was merely created by going long VFISX (or SHY) and short FXC?
That would be great!

This entry is pointing out that it would be helpful for me to have a reference so that I properly understand how a long/short portfolio is implemented, especially in the way that vincent_c is intending.

D1984, I saw in your discussions with StrategyDriven that you had some extensive data going back years before my little stash and was hoping for some excuse to ask about it. Not necessarily at the moment, but perhaps in a while, if you are open. If not, no worries.
OK, here goes:

USD to CAD historical exchange rates

https://fxtop.com/en/historical-exchang ... tnOK=Go%21

1-1-1955 to 12-31-1970; this is monthly average but if you want daily data it can give you that but you need to parse the data only a few months at a time to get that fine of data granularity; see for instance https://fxtop.com/en/historical-exchang ... tnOK=Go%21



https://fred.stlouisfed.org/series/DEXCAUS

1-1-1971 to present (daily) USD to CAD exchange rates

How many CAD each USD will buy; as the number INCREASES it means the CAD is weakening vs the USD (i.e. one USD will buy more CAD than it used to); as the number DECREASES it means the CAD is strengthening (i.e. one USD will buy less CAD than it used to). You will note the big increase from the mid-1970s to circa the very beginning of 1986 (this was when Canada had somewhat higher inflation and a weaker currency than even the USD; we in the US did have a slightly higher inflation peak in 1980 to early 1981 but overall for the period from 1974 to 1984 Canada had a bit higher inflation than the US had ....it was also--from 1979 onwards through the mid 80s--the time of the Volcker Fed, high USD interest rates, and a strong US dollar; 1986 through early 1991 was the USD losing strength vs the CAD under the Plaza Accords (an agreement to weaken the dollar against most of the other G7 currencies so as to make US exports more competitive); 1991 through late 2001 was a strengthening of the USD during the 1990s/early 2000 boom for the US vs Canada being kind of in the economic doldrums during this time (relative to the US, anyway); 2002-mid 2011 was the weakening of the USD under the Greenspan and Bernanke Fed--although do note the strengthening of the USD during the worst part of the 2008 crash; this was a time when oil/gas/coal/gold/iron/commodities/etc priced in USD soared; Canada's currency strengthened as Canada is a major natural resource and commodity producing nation; late 2011 to early 2016 was the renewed strengthening of the USD and eventually the 2014 to early 2016 commodity price collapse, and from mid-2016 to present the USD hasn't moved a whole lot against the CAD overall over the whole time period albeit it has had some blips up and down.


Just FYI: For all the exchange rate data above please do note that before the late 1960s or early 1970s the USD and CAD didn't move against each other much as this was the Bretton Woods era when currencies were virtually fixed against the US dollar and were only supposed to be allowed to move a few percent either way each year.



Canadian T-Bill yields

https://fred.stlouisfed.org/series/INTGSTCAM193N

Canadian T-bill yields (monthly average for every month since early 1950....note that these are annualized rates so you need to divide each one by 12--or maybe 12.14 or 12.15 or 12.16 or so to account for annual compounding of monthly returns when annualizing--to get actual monthly yield returns)


https://www.bankofcanada.ca/rates/inter ... ll-yields/

Canadian Treasury Bill yields daily since 1-1-2000



USD Index data

https://stooq.com/q/d/?s=dx.f&c=0

USD index monthly since 1967 (it was at 120 on 1-1-1967) and daily from 1-1-1971 onwards


https://fred.stlouisfed.org/series/DTWEXM

Trade Weighted U.S. Dollar Index (daily from 1-1-1973 to mid-2016; post mid-2016 daily data is available at https://fred.stlouisfed.org/series/DTWEXAFEGS). Note that I am not sure if UUP uses the USD index or the trade-weighted USD index (although both indexes are pretty similar and show a pretty high monthly correlation to each other) so I have included data sources for both the plain USD index and the trade-weighted USD index


US T-Bill data

https://fred.stlouisfed.org/series/TB3MS

https://fred.stlouisfed.org/series/DTB3

US 3-month T-bill rate (monthly from the mid-1930s and daily from 1-1-1954 to present)


https://fred.stlouisfed.org/series/DTB6

If for some reason you need the US 6-month T-Bill rate here it is (daily from 1-1-1958 onwards)



Finally, if you would rather just have actual monthly US T-Bill total return data rather than having the yields and having to calculate returns yourself I have that (back to the early 1940s) and can get you that as well if you'd like....just let me know.


Also, what did you wish to know about the indexes that StrategyDriven was using that helped him obtain the data for?
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Wed Jun 23, 2021 5:43 pm

vincent_c wrote:
Wed Jun 23, 2021 1:41 pm
hydromod wrote:
Tue Jun 22, 2021 8:12 am
This entry is pointing out that it would be helpful for me to have a reference so that I properly understand how a long/short portfolio is implemented, especially in the way that vincent_c is intending.
Hydromod, can you create a new thread to continue our discussion?

It's of interest to me whether using your model we can at least have another way to evaluate our portfolios. At some point, for me atleast it becomes "close enough" when a fixed allocation proxies a more complicated approach and is simpler to implement.

I'll give you an example, I have tested splitting up stocks into the fama french factors rather than just the S&P500 into the PP and rebalancing between them and it does improve the portfolio but not after fees. I have also tested diversifying exposure along the yield curve and that also improves the portfolio but it also becomes more complicated to manage.

A futures portfolio is, to me, not very difficult to manage especially compared to coming up with a quantitative approach to allocation and rebalancing. I think it all depends on investment experience and knowledge of the investment tools and I admit that when I first got into derivatives it was daunting and I had already had a pretty good theoretical understanding before actually putting money in. But once you get used to it, it's no different to buying an ETF.
I agree, a new thread would be appropriate. That way we can get a bit wonky and tutorial without polluting this thread, and we'll likely be taking stuff in a bit different direction.

I'm likely to be pretty tied up until this weekend, but then I'll have some time to devote. Saturdays are for financial learning...

I did a little bit of testing of a 3x equity balanced with a low-volatility ETF, inspired by the KBG TQQQ/BND mix. These have such extremely different volatilities that a fixed allocation seems to do better than a risk parity approach. However, starting with the QQQ/BND mix, using risk parity, and taking the average allocation gives reasonable average weights that are similar to the KGB weights once QQQ is rescaled. I would not be at all surprised if we end up with some similar findings.

I'll start a thread in tonight or tomorrow, when I can come up with an intro.
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Wed Jun 23, 2021 6:04 pm

D1984 wrote:
Tue Jun 22, 2021 11:45 pm
OK, here goes:

And how! Thank you! It looks like I'll have a fun weekend now.

Also, what did you wish to know about the indexes that StrategyDriven was using that helped him obtain the data for?
I was getting greedy about perhaps reconstructing the earlier data that StrategyDriven is using, partly to compare StrategyDriven's approaches against other approaches apples to apples with various levels of leverage. That would be a wonderful dataset.

I'm especially interested in reconstructing synthetic LETFs for some of the standard equity and treasury index funds prior to the 1990s, preferably with daily data or monthly data with monthly volatility. I want to get a better feel for how to address rising rates and shocks.

Perhaps we should follow up on this in the thread I'm starting to keep this one clean.
User avatar
StrategyDriven
Senior Member
Senior Member
Posts: 135
Joined: Thu Jul 09, 2020 5:58 pm

Re: PP Inspired Leveraged Portfolios

Post by StrategyDriven » Wed Jun 23, 2021 6:43 pm

hydromod wrote:
Wed Jun 23, 2021 6:04 pm
D1984 wrote:
Tue Jun 22, 2021 11:45 pm
OK, here goes:

And how! Thank you! It looks like I'll have a fun weekend now.

Also, what did you wish to know about the indexes that StrategyDriven was using that helped him obtain the data for?
I was getting greedy about perhaps reconstructing the earlier data that StrategyDriven is using, partly to compare StrategyDriven's approaches against other approaches apples to apples with various levels of leverage. That would be a wonderful dataset.

I'm especially interested in reconstructing synthetic LETFs for some of the standard equity and treasury index funds prior to the 1990s, preferably with daily data or monthly data with monthly volatility. I want to get a better feel for how to address rising rates and shocks.

Perhaps we should follow up on this in the thread I'm starting to keep this one clean.
I used the daily data which D1984 made available and have my own method of extending to simLETF data. What I do is take the daily 1X index data, and I take the actual LETF data since inception to current, I plot them against each other on a chart, and I use two factors for the daily 1X data, I have a leverage factor and an expense factor, I manually work with them, starting at 1.95 for 2X and 2.9 for 3X, and an expense of 3%, when I get the actual 3X and the simLETF as close as possible I call it good and then take the period of time before the actual 3X going back in time. The more math inclined out there may have better ways of using say Linear Regression, but this gets us pretty close. The fact is, I don't bank on the simLETF data to be some super close % to what it really would have been, I think it's close, but more than anything it's to get an idea how the strategies would have done and for that purpose I think it's good data.
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Wed Jun 23, 2021 8:44 pm

StrategyDriven wrote:
Wed Jun 23, 2021 6:43 pm
I used the daily data which D1984 made available and have my own method of extending to simLETF data. What I do is take the daily 1X index data, and I take the actual LETF data since inception to current, I plot them against each other on a chart, and I use two factors for the daily 1X data, I have a leverage factor and an expense factor, I manually work with them, starting at 1.95 for 2X and 2.9 for 3X, and an expense of 3%, when I get the actual 3X and the simLETF as close as possible I call it good and then take the period of time before the actual 3X going back in time. The more math inclined out there may have better ways of using say Linear Regression, but this gets us pretty close. The fact is, I don't bank on the simLETF data to be some super close % to what it really would have been, I think it's close, but more than anything it's to get an idea how the strategies would have done and for that purpose I think it's good data.
For checking return signals for strategies, that's probably good enough.

The returns will probably be optimistic outside the actual record, though, because borrowing costs were higher before. Remember, you have to borrow money to generate the excess daily returns, and interest rates were much larger back in the day. You could check the links here, especially the first one, to get more of an idea than you probably want; I don't remember the exact spot in the linked thread though. The formula is pretty straightforward, but whenever I compared it seemed like there were things going on behind the scenes that diverted from the theory. Maybe what you are doing would be better at capturing that hidden aspect.
D1984
Executive Member
Executive Member
Posts: 730
Joined: Tue Aug 16, 2011 7:23 pm

Re: PP Inspired Leveraged Portfolios

Post by D1984 » Thu Jun 24, 2021 2:07 pm

hydromod wrote:
Wed Jun 23, 2021 6:04 pm
D1984 wrote:
Tue Jun 22, 2021 11:45 pm
OK, here goes:

And how! Thank you! It looks like I'll have a fun weekend now.

Also, what did you wish to know about the indexes that StrategyDriven was using that helped him obtain the data for?
I was getting greedy about perhaps reconstructing the earlier data that StrategyDriven is using, partly to compare StrategyDriven's approaches against other approaches apples to apples with various levels of leverage. That would be a wonderful dataset.

I'm especially interested in reconstructing synthetic LETFs for some of the standard equity and treasury index funds prior to the 1990s, preferably with daily data or monthly data with monthly volatility. I want to get a better feel for how to address rising rates and shocks.

Perhaps we should follow up on this in the thread I'm starting to keep this one clean.
I'll see what I can do about getting the data reuploaded....which indexes/return series did you need and would your prefer I post the links in this thread or in the new thread you had mentioned starting?
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: PP Inspired Leveraged Portfolios

Post by Kbg » Sat Jun 26, 2021 11:23 pm

Data and discussion always welcome!

My main "ask" is to not rehash what is already in the Hedgefundie thread here.

I think discussions of gold, short vol and other diversifiers are entirely appropriate to this board including this thread and would welcome such discussion and particularly when backstopped by data.
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Sun Jul 04, 2021 1:22 pm

Without hijacking this thread any further, I just wanted to post a link to a very long entry on my Bogleheads thread where I try to wrap up my observations on the algorithm that I posted above. Comments here were quite useful in clarifying my thoughts. Now I can move on to other things...

The example compares a 1x/2x/3x version of S&P500/NASDAQ/utilities/real estate/long-term treasuries from 1986 to present (the HEDGEFUNDIE bond-bull era) running the identical algorithm for each leverage level. I think that this is a fair apples-to-apples comparison that isolates the effect of leverage. I've done similar things with just low-correlation sector funds and get similar results.

The summary is that the time-averaged portfolio is like a 60/40 allocation that swings from 80/20 to 30/70 at different times, in order to maintain a fixed risk allocation among assets. So total equity leverage with 3x LETFs swings from 0.6 to 2.4, averaging 1.8, not wildly out of line with limits based on historical observations.

I find that the CAGR is proportional to 2.2^L for this particular experiment, where L is the leverage factor, and the standard deviation is proportional to L.

The implication is that the Sharpe ratio increased substantially with increasing leverage, as does the implied optimal leverage. I think this behavior is unexpected for fixed asset allocations.

My conclusion is that dampening portfolio volatility by (i) simultaneously running several low-correlation assets (with overall positive returns) and (ii) taking advantage of volatility clustering to adjust portfolio allocations would have reduced portfolio volatility decay, thereby allowing compounding growth to be more effective over the period from 1986 to present. More effective compounding disproportionately favors greater leverage levels.

I will say that I am very dubious about growth rates persisting in the next decade or so, and adjustments likely would be necessary if the market enters an extended period of rising rates.
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: PP Inspired Leveraged Portfolios

Post by Kbg » Tue Jul 06, 2021 5:03 pm

Nice post!

Please don't bank on i or iii. Both are clearly not in the data series. On this board a generally accepted principle/assumption is we can't predict very well and that is always my assumption...but if I were to predict, current economic data suggests 1980-2020 is not in the cards for the next couple of years.
Here's rooting for TMF to maintain (i) negative correlation to equities (or at least near-zero correlation), (ii) similar volatility, and (iii) non-negative returns!
hydromod
Associate Member
Associate Member
Posts: 34
Joined: Fri Jun 18, 2021 11:55 am

Re: PP Inspired Leveraged Portfolios

Post by hydromod » Tue Jul 06, 2021 6:27 pm

Kbg wrote:
Tue Jul 06, 2021 5:03 pm
Nice post!

Please don't bank on i or iii. Both are clearly not in the data series. On this board a generally accepted principle/assumption is we can't predict very well and that is always my assumption...but if I were to predict, current economic data suggests 1980-2020 is not in the cards for the next couple of years.
Here's rooting for TMF to maintain (i) negative correlation to equities (or at least near-zero correlation), (ii) similar volatility, and (iii) non-negative returns!
I'm rooting for these TMF things because they make the ride smoother and more profitable. I'm hoping for the flight to treasury safety not to completely disappear during crashes; that's the biggie.

But more realistically, I'm expecting and trying to prepare for a choppy and perhaps net flat upcoming decade (i.e., potentially zero ten-year market returns).
Post Reply