HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Sola Scriptura
Posts: 16
Joined: Tue Feb 13, 2018 5:23 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by Sola Scriptura » Wed Jul 10, 2019 10:55 pm

I may have lost it at some point in the thread, and I tried searching in the thread to find it, but came up short:

Was there ever data presented that clearly detailed the differences in return, whether "negligible" or "significant," between monthly/quarterly/annually rebalancing of HEDGEFUNDIE's original 60 TMF/40 UPRO strategy using backtests?

Essentially, I'm interested in seeing how much return is potentially lost to annual rebalancing rather than quarterly.

Topic Author
HEDGEFUNDIE
Posts: 2791
Joined: Sun Oct 22, 2017 2:06 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by HEDGEFUNDIE » Wed Jul 10, 2019 10:57 pm

EfficientInvestor wrote:
Wed Jul 10, 2019 10:22 pm
RandomWord wrote:
Wed Jul 10, 2019 5:18 pm
robertmcd wrote:
Wed Jul 10, 2019 5:06 pm
Days like today are what worry me for this strategy, granted your stocks did well, but long term yields went up 3 bps while short term went down 10 bps. Somebody please come out with a leveraged 2 yr treasury ETF so I can run a risk parity strategy with it.
If you're willing to use futures instead of an ETF, the futures market has this covered.

But I'm not sure how much difference it makes in the long term, seems like the sort of thing that flucuates from day to day but overall averages out.
I started implementing the strategy with futures contracts a few weeks ago. I'm using a Roth IRA account that has about $32k in it. I have it spread across 2 contracts of /MES (micro S&P), 1 contract of /ZT (2-year treasuries), and 1 contract of /MGC (micro gold). This roughly equates to a 15/80/5 portfolio leveraged by a factor of 8. I have about $4k that is covering the margin requirements and I keep another $2k or so as cash in order to cover any daily fluctuations. The remaining cash is in MINT (ultra-short term bond ETF) to help offset some of the borrowing cost associated with the futures. The backtest below (since Nov 1991) represents what I'm aiming for. I'm assuming a quarterly rebalance since I will do the rebalance every quarter when I need to roll the futures contracts. In times when I may need 2.5 contracts of /MES or 1.5 contracts of /ZT, I will likely supplement the futures with regular ETFs (SPY, SHY, GLD) in order to hit my target exposures and maintain an 8X 15/80/5 portfolio.

https://www.portfoliovisualizer.com/bac ... 0&total3=0

This is my first venture into the use of futures, so I'm still learning some of the intricacies. However, my hope is that this will be a better alternative than the leveraged ETFs because it will help me avoid the 1% expense ratios and avoid the volatility drag due to the daily leverage reset. That drag could add up to another 1.5-2% over time. Therefore, I think I could conservatively expect a 2% increase in return per year over using leveraged ETFs. In addition, the futures allow me to use shorter term bonds that help limit my interest rate risk. We shall see how it goes.
Very cool. I suggest you keep this thread updated on your progress quarterly as you roll new contracts.

samsdad
Posts: 671
Joined: Sat Jan 02, 2016 6:20 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by samsdad » Thu Jul 11, 2019 12:08 am

Sola Scriptura wrote:
Wed Jul 10, 2019 10:55 pm
I may have lost it at some point in the thread, and I tried searching in the thread to find it, but came up short:

Was there ever data presented that clearly detailed the differences in return, whether "negligible" or "significant," between monthly/quarterly/annually rebalancing of HEDGEFUNDIE's original 60 TMF/40 UPRO strategy using backtests?

Essentially, I'm interested in seeing how much return is potentially lost to annual rebalancing rather than quarterly.
UtiliIizing the HEDGEFUNDIE data linked in the OP:

Jan 1987 - Jan 2019
No rebalancing: 10.66% CAGR
Annual: 16.21%
Semi-annual: 15.96%
Quarterly: 17.01%
Monthly: 15.29%

For reference: the Vanguard 500 index investor shares returned 10.11% over the same period.

The annual rebalancer would’ve ended up with 1.2 million nominal, the quarterly rebalancer 1.5 million nominal.

MotoTrojan, please double check my math—I’m on my mobile again :shock:
Last edited by samsdad on Thu Jul 11, 2019 12:16 am, edited 1 time in total.

Hydromod
Posts: 88
Joined: Tue Mar 26, 2019 10:21 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by Hydromod » Thu Jul 11, 2019 12:13 am

Sola Scriptura wrote:
Wed Jul 10, 2019 10:55 pm
I may have lost it at some point in the thread, and I tried searching in the thread to find it, but came up short:

Was there ever data presented that clearly detailed the differences in return, whether "negligible" or "significant," between monthly/quarterly/annually rebalancing of HEDGEFUNDIE's original 60 TMF/40 UPRO strategy using backtests?

Essentially, I'm interested in seeing how much return is potentially lost to annual rebalancing rather than quarterly.
I tried looking at this issue in a companion thread, refinements to hedgefundie's excellent approach.

Basically the results become more sensitive to start date as the rebalancing frequency becomes longer. You get more of a spread in outcomes, both favorable and unfavorable, and the sequences become more volatile. You also seem to lose a little return on average.

With that said, there is not much difference between monthly and quarterly. The semiannual and annual cases start to get noisier.

I would link if I wasn't fumbling around with an iPad.

MoneyMarathon
Posts: 243
Joined: Sun Sep 30, 2012 3:38 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MoneyMarathon » Thu Jul 11, 2019 12:20 am

In the Logical Invest article here: https://logical-invest.com/leveraged-un ... -strategy/

I noticed this: "Note the Maximum DrawDown of -94% (-66.39%) in the synthetic SPXL (TMF [sic?]) from Oct 2008 to March 2009. In practical terms, $100 invested in SPXL would have decreased to $6."

However, using the UPROSIM data in portfolio visualizer (shared in OP), I get a drawdown -83.15%, i.e. from $100 to about $17. Which gives you about 3x the money left in the stock portion of the portfolio (if not already rebalanced in the meantime).

Edit: I may have figured out the discrepancy. The article may have misstated the time period. The drawdown in Portfolio Visualizer is indeed -94.33% if you start from the high in June 2007 down to the low in March 2009. The drawdown from the high in 99 to 09 is even more wild, -97.48%. Good thing it's just one part of a well-balanced portfolio. :)

The max drawdown of TMFSIM is just behind us, from August 2016 to October 2018, at -47.1%. The max drawdown of TMF itself was from August 2012 to December 2013, at -49.74%.

samsdad
Posts: 671
Joined: Sat Jan 02, 2016 6:20 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by samsdad » Thu Jul 11, 2019 12:39 am

MoneyMarathon wrote:
Thu Jul 11, 2019 12:20 am
In the Logical Invest article here: https://logical-invest.com/leveraged-un ... -strategy/

I noticed this: "Note the Maximum DrawDown of -94% (-66.39%) in the synthetic SPXL (TMF [sic?]) from Oct 2008 to March 2009. In practical terms, $100 invested in SPXL would have decreased to $6."

However, using the UPROSIM data in portfolio visualizer (shared in OP), I get this: max drawdown -83.15%, i.e. from $100 to about $17. Which gives you about 3x the money left in the stock portion of the portfolio (if not already rebalanced in the meantime), a possibly-significant difference.

Does anyone understand the discrepancy? Are there aspects of path dependence that might not be accounted for in UPROSIM data? How is the UPROSIM data modeling the effect of volatility and changes in the volatility of SPY (if it is)?

Does anyone know the true best estimate for a max drawdown in that time period for a 100% UPRO allocation?

Apologies if this has all been answered already.
I get the same drawdown from the OP data (-83.15%). The 1955+ data gives -83.11%. PV says if you started with $100, you’d still have $20.42 or $20.46, respectively, or about 20% more than the $17 you think you’d have left over—so look on the bright side.

Perhaps the authors are wrong?

Edit: If you start from a year earlier, October 2007, you get -94.14% with the OP data, and -94.15% with the 1955+. However, you’d still have $7.31, not the $6 as claimed (about 22% more) :o.

Edit 2: I see you saw the date discrepancy too.
Last edited by samsdad on Thu Jul 11, 2019 12:41 am, edited 1 time in total.

MoneyMarathon
Posts: 243
Joined: Sun Sep 30, 2012 3:38 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MoneyMarathon » Thu Jul 11, 2019 12:41 am

Perhaps the authors are wrong?
I think so! Edited my post. :sharebeer

MoneyMarathon
Posts: 243
Joined: Sun Sep 30, 2012 3:38 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MoneyMarathon » Thu Jul 11, 2019 4:31 am

I've been working on getting a better understanding of the nature of "volatility decay." Rebalancing interacts a lot.

Here is 100% SDY (blue) and 50% SDYL, 50% CASHX (red) with no rebalancing. SDYL is a monthly resetting leveraged 2x ETF.

Image

Notice that the SDYL portfolio is shifting its allocation over time to have more SDYL than the original 50%. This increases its exposure to the underlying, over time, to be greater than 1x. Basically, beta is drifting up over time, and as the beta increases, the portfolio does what you'd expect: further gains are magnified, but so are subsequent losses.

That is, until you rebalance. At that point you can reset the portfolio to have 1x exposure by making the 2x part be 50% again. The only issue still remaining here is that you waited to reset the portfolio, so you've already drifted away from your underlying due to the already-deviated returns of the previous 11 months in which you had not yet rebalanced. With annual rebalancing, the two portfolios are much closer.

Image

Because the path dependency of returns can work for you or against you, the annual-rebalanced portfolio here gains and loses several times compared to its underlying. But once you rebalance monthly, with a monthly-resetting 2x ETF at 50% weight, the tracking error of the ETF itself is the only remaining issue.

Image

How does it track so well now? The basic idea in this case is that you take about half of the cash off the table when you win (and your next bet would naturally be bigger), but you put about half of what was lost back in when you lose (and your next bet would naturally be smaller). This is intended to keep a relatively-constant level of exposure in the next bet (it's not a prediction, no reversion is really expected).

Obviously this requires a source of funds from which to play this side game of adding and removing funds from your primary bet. The ideal source would both (a) have its own rate of return on otherwise idle cash and (b) be there in full when needed to get back in after a big loss. The safest way to esnure (b) is to have the side game be a large-enough fraction of funds and low-enough volatility, so that pulling enough money from the lower-yielding side game is possible. However, the ideal way to do it is to have a side game that (a) gives the same size of wins as the main bet has losses and (b) gives wins when the main bet has losses. This is basically a convoluted way of explaining some of the ideas of modern portfolio theory and risk parity.

Modern portfolio theory was invented long before we had leveraged ETFs. The same ideas that are used to mitigate "sequence of returns" risk are also used to reduce the effects of "volatility decay" with leveraged ETFs. Rebalancing is crucial to any long-term use of leveraged ETFs.

MotoTrojan
Posts: 5011
Joined: Wed Feb 01, 2017 8:39 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MotoTrojan » Thu Jul 11, 2019 7:55 am

MoneyMarathon wrote:
Thu Jul 11, 2019 4:31 am
I've been working on getting a better understanding of the nature of "volatility decay." Rebalancing interacts a lot.

Here is 100% SDY (blue) and 50% SDYL, 50% CASHX (red) with no rebalancing. SDYL is a monthly resetting leveraged 2x ETF.

Image

Notice that the SDYL portfolio is shifting its allocation over time to have more SDYL than the original 50%. This increases its exposure to the underlying, over time, to be greater than 1x. Basically, beta is drifting up over time, and as the beta increases, the portfolio does what you'd expect: further gains are magnified, but so are subsequent losses.

That is, until you rebalance. At that point you can reset the portfolio to have 1x exposure by making the 2x part be 50% again. The only issue still remaining here is that you waited to reset the portfolio, so you've already drifted away from your underlying due to the already-deviated returns of the previous 11 months in which you had not yet rebalanced. With annual rebalancing, the two portfolios are much closer.

Image

Because the path dependency of returns can work for you or against you, the annual-rebalanced portfolio here gains and loses several times compared to its underlying. But once you rebalance monthly, with a monthly-resetting 2x ETF at 50% weight, the tracking error of the ETF itself is the only remaining issue.

Image

How does it track so well now? The basic idea in this case is that you take about half of the cash off the table when you win (and your next bet would naturally be bigger), but you put about half of what was lost back in when you lose (and your next bet would naturally be smaller). This is intended to keep a relatively-constant level of exposure in the next bet (it's not a prediction, no reversion is really expected).

Obviously this requires a source of funds from which to play this side game of adding and removing funds from your primary bet. The ideal source would both (a) have its own rate of return on otherwise idle cash and (b) be there in full when needed to get back in after a big loss. The safest way to esnure (b) is to have the side game be a large-enough fraction of funds and low-enough volatility, so that pulling enough money from the lower-yielding side game is possible. However, the ideal way to do it is to have a side game that (a) gives the same size of wins as the main bet has losses and (b) gives wins when the main bet has losses. This is basically a convoluted way of explaining some of the ideas of modern portfolio theory and risk parity.

Modern portfolio theory was invented long before we had leveraged ETFs. The same ideas that are used to mitigate "sequence of returns" risk are also used to reduce the effects of "volatility decay" with leveraged ETFs. Rebalancing is crucial to any long-term use of leveraged ETFs.
Frankly I don't think this analysis is robust and I haven't seen any explanation for how you handle/explain volatility decay. Are you sure you are thinking about volatility decay correctly? Example: Index EX goes from 1 to 1.1 and back to 1 and repeats that forever. What will happen to an index that returns 3x the daily return of EX? It will slowly decay in value, even though the base index isn't going anywhere. Rebalancing can't change that, unless you rebalance at the same frequency as the ETF (daily in the case of these funds).

robertmcd
Posts: 436
Joined: Tue Aug 09, 2016 9:06 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by robertmcd » Thu Jul 11, 2019 9:20 am

Clarice wrote:
Wed Jul 10, 2019 9:54 pm
MotoTrojan wrote:
Wed Jul 10, 2019 7:47 pm
Clarice wrote:
Wed Jul 10, 2019 7:43 pm
Can I just ask - and sorry if it has been asked already -

If this strategy returns a +5% on one particular day, is there a collective -5% return for anyone else?

Struggling to understand who is taking the other side and if it is equal sum.

Thank you!
If the S&P500 returns 5% in one particular day, is there a collective -5% return for everyone else? Your index funds use derivative products to meet a small amount of their index exposure too so they have some cash for redemption etc., this is similar but just blown up to a larger level.
I’m not sure I understand. If S&P500 is up 5% on one particular day, then those long the S&P earned 5% and those short the S&P lost 5%. If no one is short, then everyone gained 5% and vice versa. Please correct me if I’m mistaken.

In this case, I am wondering if the same holds true.
I feel like there’s 1 short for every 1 long for this strategy. Although admittedly not sure why I’m getting myself so hung up!
Another way to look at it is that this strategy depends on stocks and bonds outperforming cash/T-bills. You are effectively shorting T bills relative to stocks and treasury bonds. The historical negative correlation between stocks and treasury bonds, along with their expected/historical return over cash, is what allows for the use of more leverage than one could do if investing in either in isolation.

unknownfuture
Posts: 27
Joined: Fri Feb 15, 2019 3:30 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by unknownfuture » Thu Jul 11, 2019 11:51 am

EfficientInvestor wrote:
Wed Jul 10, 2019 10:22 pm
RandomWord wrote:
Wed Jul 10, 2019 5:18 pm
robertmcd wrote:
Wed Jul 10, 2019 5:06 pm
Days like today are what worry me for this strategy, granted your stocks did well, but long term yields went up 3 bps while short term went down 10 bps. Somebody please come out with a leveraged 2 yr treasury ETF so I can run a risk parity strategy with it.
If you're willing to use futures instead of an ETF, the futures market has this covered.

But I'm not sure how much difference it makes in the long term, seems like the sort of thing that flucuates from day to day but overall averages out.
I started implementing the strategy with futures contracts a few weeks ago. I'm using a Roth IRA account that has about $32k in it. I have it spread across 2 contracts of /MES (micro S&P), 1 contract of /ZT (2-year treasuries), and 1 contract of /MGC (micro gold). This roughly equates to a 15/80/5 portfolio leveraged by a factor of 8. I have about $4k that is covering the margin requirements and I keep another $2k or so as cash in order to cover any daily fluctuations. The remaining cash is in MINT (ultra-short term bond ETF) to help offset some of the borrowing cost associated with the futures. The backtest below (since Nov 1991) represents what I'm aiming for. I'm assuming a quarterly rebalance since I will do the rebalance every quarter when I need to roll the futures contracts. In times when I may need 2.5 contracts of /MES or 1.5 contracts of /ZT, I will likely supplement the futures with regular ETFs (SPY, SHY, GLD) in order to hit my target exposures and maintain an 8X 15/80/5 portfolio.

https://www.portfoliovisualizer.com/bac ... 0&total3=0

This is my first venture into the use of futures, so I'm still learning some of the intricacies. However, my hope is that this will be a better alternative than the leveraged ETFs because it will help me avoid the 1% expense ratios and avoid the volatility drag due to the daily leverage reset. That drag could add up to another 1.5-2% over time. Therefore, I think I could conservatively expect a 2% increase in return per year over using leveraged ETFs. In addition, the futures allow me to use shorter term bonds that help limit my interest rate risk. We shall see how it goes.
Wow, amazing! Better returns, lower variance, and (best of all) this strategy uses 2-year treasuries, therefore much less sensitive to changes in interest rates (over long enough time periods).

If I understand correctly, in a taxable account you'd be fully realizing capital gains ever quarter/year, and those for futures contracts are taxed as 40% short-term 60% long-term capital gains. That would take a serious bite out of profits (especially in California).

RandomWord
Posts: 18
Joined: Tue Jun 18, 2019 1:12 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by RandomWord » Thu Jul 11, 2019 12:12 pm

unknownfuture wrote:
Thu Jul 11, 2019 11:51 am
EfficientInvestor wrote:
Wed Jul 10, 2019 10:22 pm
RandomWord wrote:
Wed Jul 10, 2019 5:18 pm
robertmcd wrote:
Wed Jul 10, 2019 5:06 pm
Days like today are what worry me for this strategy, granted your stocks did well, but long term yields went up 3 bps while short term went down 10 bps. Somebody please come out with a leveraged 2 yr treasury ETF so I can run a risk parity strategy with it.
If you're willing to use futures instead of an ETF, the futures market has this covered.

But I'm not sure how much difference it makes in the long term, seems like the sort of thing that flucuates from day to day but overall averages out.
I started implementing the strategy with futures contracts a few weeks ago. I'm using a Roth IRA account that has about $32k in it. I have it spread across 2 contracts of /MES (micro S&P), 1 contract of /ZT (2-year treasuries), and 1 contract of /MGC (micro gold). This roughly equates to a 15/80/5 portfolio leveraged by a factor of 8. I have about $4k that is covering the margin requirements and I keep another $2k or so as cash in order to cover any daily fluctuations. The remaining cash is in MINT (ultra-short term bond ETF) to help offset some of the borrowing cost associated with the futures. The backtest below (since Nov 1991) represents what I'm aiming for. I'm assuming a quarterly rebalance since I will do the rebalance every quarter when I need to roll the futures contracts. In times when I may need 2.5 contracts of /MES or 1.5 contracts of /ZT, I will likely supplement the futures with regular ETFs (SPY, SHY, GLD) in order to hit my target exposures and maintain an 8X 15/80/5 portfolio.

https://www.portfoliovisualizer.com/bac ... 0&total3=0

This is my first venture into the use of futures, so I'm still learning some of the intricacies. However, my hope is that this will be a better alternative than the leveraged ETFs because it will help me avoid the 1% expense ratios and avoid the volatility drag due to the daily leverage reset. That drag could add up to another 1.5-2% over time. Therefore, I think I could conservatively expect a 2% increase in return per year over using leveraged ETFs. In addition, the futures allow me to use shorter term bonds that help limit my interest rate risk. We shall see how it goes.
Wow, amazing! Better returns, lower variance, and (best of all) this strategy uses 2-year treasuries, therefore much less sensitive to changes in interest rates (over long enough time periods).

If I understand correctly, in a taxable account you'd be fully realizing capital gains ever quarter/year, and those for futures contracts are taxed as 40% short-term 60% long-term capital gains. That would take a serious bite out of profits (especially in California).
That's essentially what I've been doing for the past few years. And yeah, taxes are annoying. But the good news is that you only have to pay taxes if it goes up, and if it performs like it did in backtests (and this year), the performance is so amazing that I wouldn't worry too much about taxes. If it crashes, at least you get a deduction.

Taxes for treasury futures are arguably better than on holding bonds directly, since the bond payments would otherwise be all taxed at ordinary income tax rates.

MoneyMarathon
Posts: 243
Joined: Sun Sep 30, 2012 3:38 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MoneyMarathon » Thu Jul 11, 2019 1:02 pm

MotoTrojan wrote:
Thu Jul 11, 2019 7:55 am
Frankly I don't think this analysis is robust
I was going for easy to understand, not robust.
MotoTrojan wrote:
Thu Jul 11, 2019 7:55 am
Example: Index EX goes from 1 to 1.1 and back to 1 and repeats that forever. What will happen to an index that returns 3x the daily return of EX? It will slowly decay in value, even though the base index isn't going anywhere.
Right.
MotoTrojan wrote:
Thu Jul 11, 2019 7:55 am
Rebalancing can't change that, unless you rebalance at the same frequency as the ETF (daily in the case of these funds).
If loading up on beta (stock exposure) doesn't pay in the long run, of course, you'd end up net negative if the rebalancing isn't executed perfectly and at the same cadence as the reset in leverage, as you point out. And you'll still end up negative due to expenses and trading costs, even then.

In practice, on real world data, rebalancing can mitigate the effects of volatility decay. You can get closer to the underlying by rebalancing more frequently. You can see this by setting the cash account to 66% and rebalancing for a 3x ETF. The portfolio will usually track 1x exposure to the underlying better when rebalancing is more frequent, terribly when rebalancing isn't done at all.

Most examples of volatility decay show 100% exposure to the leveraged ETF, so they don't show the effects of rebalancing. I just wanted to give an example of how rebalancing interacts.

EfficientInvestor
Posts: 155
Joined: Thu Nov 01, 2018 7:02 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by EfficientInvestor » Thu Jul 11, 2019 1:11 pm

unknownfuture wrote:
Thu Jul 11, 2019 11:51 am
Wow, amazing! Better returns, lower variance, and (best of all) this strategy uses 2-year treasuries, therefore much less sensitive to changes in interest rates (over long enough time periods).

If I understand correctly, in a taxable account you'd be fully realizing capital gains ever quarter/year, and those for futures contracts are taxed as 40% short-term 60% long-term capital gains. That would take a serious bite out of profits (especially in California).
If applying strategy in taxable account, you could consider using options on ETFs that have an expiration at least 1 year in future (also known as LEAPS). That way you can get taxed at long term capital gains rates as long as you only rebalance once per year. SPY options are fairly liquid that far out and far enough in the money to not have to pay too much extrinsic value premium. LEAPS on the bond ETFs are less liquid and probably more susceptible to inefficiencies due to bid/ask spread. The other benefit of the LEAPS over futures is you have insurance built in (can’t lose more than you put in) and you don’t have to worry about margin calls. I intend to start using this strategy at some point in a taxable account for the purpose of trying it out. The other benefit is you don’t have to have as large of an account as compared to futures.

dalbright
Posts: 148
Joined: Sun Oct 21, 2018 6:23 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by dalbright » Thu Jul 11, 2019 1:21 pm

Finally going to join in here on the adventure shortly, albeit at a much smaller level than most of you! I'm still working on building up my less volatile "normal" roth 401k/IRA retirement accounts, but since no leveraged funds are allowed in there I just opened up a m1 finance account for this purpose. Will be handy since I will only be adding 50-100 biweekly, thus saves on trading fees which would nullify such small adds. Hopefully someday will be able to add in more :).

MotoTrojan
Posts: 5011
Joined: Wed Feb 01, 2017 8:39 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MotoTrojan » Thu Jul 11, 2019 3:10 pm

MoneyMarathon wrote:
Thu Jul 11, 2019 1:02 pm
MotoTrojan wrote:
Thu Jul 11, 2019 7:55 am
Frankly I don't think this analysis is robust
I was going for easy to understand, not robust.
MotoTrojan wrote:
Thu Jul 11, 2019 7:55 am
Example: Index EX goes from 1 to 1.1 and back to 1 and repeats that forever. What will happen to an index that returns 3x the daily return of EX? It will slowly decay in value, even though the base index isn't going anywhere.
Right.
MotoTrojan wrote:
Thu Jul 11, 2019 7:55 am
Rebalancing can't change that, unless you rebalance at the same frequency as the ETF (daily in the case of these funds).
If loading up on beta (stock exposure) doesn't pay in the long run, of course, you'd end up net negative if the rebalancing isn't executed perfectly and at the same cadence as the reset in leverage, as you point out. And you'll still end up negative due to expenses and trading costs, even then.

In practice, on real world data, rebalancing can mitigate the effects of volatility decay. You can get closer to the underlying by rebalancing more frequently. You can see this by setting the cash account to 66% and rebalancing for a 3x ETF. The portfolio will usually track 1x exposure to the underlying better when rebalancing is more frequent, terribly when rebalancing isn't done at all.

Most examples of volatility decay show 100% exposure to the leveraged ETF, so they don't show the effects of rebalancing. I just wanted to give an example of how rebalancing interacts.
Gotcha that totally makes sense I see your point now; it isn't binary and the more frequent you rebalance the less volatility decay impacts you. Thanks.

CulchaCity
Posts: 6
Joined: Sun Jul 02, 2017 5:28 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by CulchaCity » Thu Jul 11, 2019 6:02 pm

EfficientInvestor wrote:
Wed Jul 10, 2019 10:22 pm
RandomWord wrote:
Wed Jul 10, 2019 5:18 pm
robertmcd wrote:
Wed Jul 10, 2019 5:06 pm
Days like today are what worry me for this strategy, granted your stocks did well, but long term yields went up 3 bps while short term went down 10 bps. Somebody please come out with a leveraged 2 yr treasury ETF so I can run a risk parity strategy with it.
If you're willing to use futures instead of an ETF, the futures market has this covered.

But I'm not sure how much difference it makes in the long term, seems like the sort of thing that flucuates from day to day but overall averages out.
I started implementing the strategy with futures contracts a few weeks ago. I'm using a Roth IRA account that has about $32k in it. I have it spread across 2 contracts of /MES (micro S&P), 1 contract of /ZT (2-year treasuries), and 1 contract of /MGC (micro gold). This roughly equates to a 15/80/5 portfolio leveraged by a factor of 8. I have about $4k that is covering the margin requirements and I keep another $2k or so as cash in order to cover any daily fluctuations. The remaining cash is in MINT (ultra-short term bond ETF) to help offset some of the borrowing cost associated with the futures. The backtest below (since Nov 1991) represents what I'm aiming for. I'm assuming a quarterly rebalance since I will do the rebalance every quarter when I need to roll the futures contracts. In times when I may need 2.5 contracts of /MES or 1.5 contracts of /ZT, I will likely supplement the futures with regular ETFs (SPY, SHY, GLD) in order to hit my target exposures and maintain an 8X 15/80/5 portfolio.

https://www.portfoliovisualizer.com/bac ... 0&total3=0

This is my first venture into the use of futures, so I'm still learning some of the intricacies. However, my hope is that this will be a better alternative than the leveraged ETFs because it will help me avoid the 1% expense ratios and avoid the volatility drag due to the daily leverage reset. That drag could add up to another 1.5-2% over time. Therefore, I think I could conservatively expect a 2% increase in return per year over using leveraged ETFs. In addition, the futures allow me to use shorter term bonds that help limit my interest rate risk. We shall see how it goes.
What brokerage are you using to trade futures contracts in a Roth IRA?
Thanks

EfficientInvestor
Posts: 155
Joined: Thu Nov 01, 2018 7:02 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by EfficientInvestor » Thu Jul 11, 2019 8:23 pm

CulchaCity wrote:
Thu Jul 11, 2019 6:02 pm
EfficientInvestor wrote:
Wed Jul 10, 2019 10:22 pm
RandomWord wrote:
Wed Jul 10, 2019 5:18 pm
robertmcd wrote:
Wed Jul 10, 2019 5:06 pm
Days like today are what worry me for this strategy, granted your stocks did well, but long term yields went up 3 bps while short term went down 10 bps. Somebody please come out with a leveraged 2 yr treasury ETF so I can run a risk parity strategy with it.
If you're willing to use futures instead of an ETF, the futures market has this covered.

But I'm not sure how much difference it makes in the long term, seems like the sort of thing that flucuates from day to day but overall averages out.
I started implementing the strategy with futures contracts a few weeks ago. I'm using a Roth IRA account that has about $32k in it. I have it spread across 2 contracts of /MES (micro S&P), 1 contract of /ZT (2-year treasuries), and 1 contract of /MGC (micro gold). This roughly equates to a 15/80/5 portfolio leveraged by a factor of 8. I have about $4k that is covering the margin requirements and I keep another $2k or so as cash in order to cover any daily fluctuations. The remaining cash is in MINT (ultra-short term bond ETF) to help offset some of the borrowing cost associated with the futures. The backtest below (since Nov 1991) represents what I'm aiming for. I'm assuming a quarterly rebalance since I will do the rebalance every quarter when I need to roll the futures contracts. In times when I may need 2.5 contracts of /MES or 1.5 contracts of /ZT, I will likely supplement the futures with regular ETFs (SPY, SHY, GLD) in order to hit my target exposures and maintain an 8X 15/80/5 portfolio.

https://www.portfoliovisualizer.com/bac ... 0&total3=0

This is my first venture into the use of futures, so I'm still learning some of the intricacies. However, my hope is that this will be a better alternative than the leveraged ETFs because it will help me avoid the 1% expense ratios and avoid the volatility drag due to the daily leverage reset. That drag could add up to another 1.5-2% over time. Therefore, I think I could conservatively expect a 2% increase in return per year over using leveraged ETFs. In addition, the futures allow me to use shorter term bonds that help limit my interest rate risk. We shall see how it goes.
What brokerage are you using to trade futures contracts in a Roth IRA?
Thanks
TastyWorks. They are really ahead of the game in helping individual investors have better access to more advanced products. They may be the only broker that allows you to trade options on futures in an IRA. They also have great education via their TastyTrade platform and are extremely competitive in their pricing structure.

User avatar
mrspock
Posts: 402
Joined: Tue Feb 13, 2018 2:49 am
Location: Vulcan

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by mrspock » Thu Jul 11, 2019 8:45 pm

I wonder if M1 is scratching their heads at all this TMF & UPRO they now have in their brokerage :D . Nevermind in 20 years....

dziuniek
Posts: 571
Joined: Mon Jul 23, 2012 2:54 pm
Location: Corrupticut

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by dziuniek » Thu Jul 11, 2019 8:48 pm

mrspock wrote:
Thu Jul 11, 2019 8:45 pm
I wonder if M1 is scratching their heads at all this TMF & UPRO they now have in their brokerage :D . Nevermind in 20 years....
Haha, quite possibly :)

User avatar
willthrill81
Posts: 11041
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by willthrill81 » Thu Jul 11, 2019 8:57 pm

mrspock wrote:
Thu Jul 11, 2019 8:45 pm
I wonder if M1 is scratching their heads at all this TMF & UPRO they now have in their brokerage :D . Nevermind in 20 years....
Very possibly. I also wonder if ProShares and Direxion have noticed an uptick in their overnight AUM.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
Dr. Long
Posts: 19
Joined: Sun Jun 23, 2019 2:14 pm
Location: Location, Location, Location

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by Dr. Long » Thu Jul 11, 2019 9:06 pm

I really like M1. I currently have 20% of my total portfolio with them - and a quarter of that (5%) devoted to this delightful strategy (experiment?)!
"(It's) the economy, stupid," - James Carville

Dockpp
Posts: 6
Joined: Mon Jun 06, 2016 10:21 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by Dockpp » Thu Jul 11, 2019 9:48 pm

RandomWord wrote:
Thu Jul 11, 2019 12:12 pm
unknownfuture wrote:
Thu Jul 11, 2019 11:51 am
EfficientInvestor wrote:
Wed Jul 10, 2019 10:22 pm
RandomWord wrote:
Wed Jul 10, 2019 5:18 pm
robertmcd wrote:
Wed Jul 10, 2019 5:06 pm
Days like today are what worry me for this strategy, granted your stocks did well, but long term yields went up 3 bps while short term went down 10 bps. Somebody please come out with a leveraged 2 yr treasury ETF so I can run a risk parity strategy with it.
If you're willing to use futures instead of an ETF, the futures market has this covered.

But I'm not sure how much difference it makes in the long term, seems like the sort of thing that flucuates from day to day but overall averages out.
I started implementing the strategy with futures contracts a few weeks ago. I'm using a Roth IRA account that has about $32k in it. I have it spread across 2 contracts of /MES (micro S&P), 1 contract of /ZT (2-year treasuries), and 1 contract of /MGC (micro gold). This roughly equates to a 15/80/5 portfolio leveraged by a factor of 8. I have about $4k that is covering the margin requirements and I keep another $2k or so as cash in order to cover any daily fluctuations. The remaining cash is in MINT (ultra-short term bond ETF) to help offset some of the borrowing cost associated with the futures. The backtest below (since Nov 1991) represents what I'm aiming for. I'm assuming a quarterly rebalance since I will do the rebalance every quarter when I need to roll the futures contracts. In times when I may need 2.5 contracts of /MES or 1.5 contracts of /ZT, I will likely supplement the futures with regular ETFs (SPY, SHY, GLD) in order to hit my target exposures and maintain an 8X 15/80/5 portfolio.

https://www.portfoliovisualizer.com/bac ... 0&total3=0

This is my first venture into the use of futures, so I'm still learning some of the intricacies. However, my hope is that this will be a better alternative than the leveraged ETFs because it will help me avoid the 1% expense ratios and avoid the volatility drag due to the daily leverage reset. That drag could add up to another 1.5-2% over time. Therefore, I think I could conservatively expect a 2% increase in return per year over using leveraged ETFs. In addition, the futures allow me to use shorter term bonds that help limit my interest rate risk. We shall see how it goes.
Wow, amazing! Better returns, lower variance, and (best of all) this strategy uses 2-year treasuries, therefore much less sensitive to changes in interest rates (over long enough time periods).

If I understand correctly, in a taxable account you'd be fully realizing capital gains ever quarter/year, and those for futures contracts are taxed as 40% short-term 60% long-term capital gains. That would take a serious bite out of profits (especially in California).
That's essentially what I've been doing for the past few years. And yeah, taxes are annoying. But the good news is that you only have to pay taxes if it goes up, and if it performs like it did in backtests (and this year), the performance is so amazing that I wouldn't worry too much about taxes. If it crashes, at least you get a deduction.

Taxes for treasury futures are arguably better than on holding bonds directly, since the bond payments would otherwise be all taxed at ordinary income tax rates.
Where do dividend fit in as the future do not pay any?

MotoTrojan
Posts: 5011
Joined: Wed Feb 01, 2017 8:39 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MotoTrojan » Thu Jul 11, 2019 9:54 pm

Dockpp wrote:
Thu Jul 11, 2019 9:48 pm

Where do dividend fit in as the future do not pay any?
They don't. But the savings on ER, borrowing rate, and volatility drag could come out ahead. I say could because daily rebalancing (volatility drag) can also create outsized returns (>3x the base asset).

EfficientInvestor
Posts: 155
Joined: Thu Nov 01, 2018 7:02 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by EfficientInvestor » Thu Jul 11, 2019 10:22 pm

MotoTrojan wrote:
Thu Jul 11, 2019 9:54 pm
Dockpp wrote:
Thu Jul 11, 2019 9:48 pm

Where do dividend fit in as the future do not pay any?
They don't. But the savings on ER, borrowing rate, and volatility drag could come out ahead. I say could because daily rebalancing (volatility drag) can also create outsized returns (>3x the base asset).
Dividends are accounted for in the pricing of the futures contact. The dividends that are expected over the life of the contract are subtracted from the spot price of the underlying. In other words...you still get the benefit of the dividend, but the benefit is received up front via the discounted purchase price. If this weren’t the case, an arbitrage opportunity would exist in which you could go long the underlying, short the future, and pocket the dividend with zero risk. See the “Forward Price Formula” at the wiki page below.

https://en.m.wikipedia.org/wiki/Forward ... prov=sfti1

MotoTrojan
Posts: 5011
Joined: Wed Feb 01, 2017 8:39 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MotoTrojan » Thu Jul 11, 2019 11:08 pm

EfficientInvestor wrote:
Thu Jul 11, 2019 10:22 pm
MotoTrojan wrote:
Thu Jul 11, 2019 9:54 pm
Dockpp wrote:
Thu Jul 11, 2019 9:48 pm

Where do dividend fit in as the future do not pay any?
They don't. But the savings on ER, borrowing rate, and volatility drag could come out ahead. I say could because daily rebalancing (volatility drag) can also create outsized returns (>3x the base asset).
Dividends are accounted for in the pricing of the futures contact. The dividends that are expected over the life of the contract are subtracted from the spot price of the underlying. In other words...you still get the benefit of the dividend, but the benefit is received up front via the discounted purchase price. If this weren’t the case, an arbitrage opportunity would exist in which you could go long the underlying, short the future, and pocket the dividend with zero risk. See the “Forward Price Formula” at the wiki page below.

https://en.m.wikipedia.org/wiki/Forward ... prov=sfti1
Thanks! I must be thinking about LEAPS?

MoneyMarathon
Posts: 243
Joined: Sun Sep 30, 2012 3:38 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MoneyMarathon » Fri Jul 12, 2019 2:45 am

Others have made similar comments, but it's worth noting again how different the 1987-2018 data in Portfolio Visualizer is from the 1955-2018 data in Simba's spreadsheet. This is true even if you want to go with just TMF (3x LLT) and UPRO (3x SP500). It seems like the 60 TMF / 40 UPRO balance came from doing portfolio optimization: risk parity and maximizing Sharpe each give about this ratio... on the 1987-2018 data.

Long bonds did well enough during this period that it hardly mattered much (especially without knowing the precise results beforehand, which nobody could know...) whether you were 40/60 favoring bonds, 50/50, or 60/40 favoring stocks. They all had 16-17% CAGR in this time period.

Image

The story's very different in the 1955-2018 data. Then it very much matters for how many millions you end up with.

Image

With CAGR of 11.65%, 12.63%, and 13.25% respectively... where more stocks (up to 60% anyway) meant higher returns.

Standard deviation of 31.24%, 32.48%, and 34.96%. Sharpe of 0.35, 0.38, and 0.39 respectively. Increasing Sharpe and higher CAGR.

You'd have to give up on the idea that the graph will be near-flat in a bear market, but that depends a little on the luck of timing for rebalancing in addition to betting on the negative correlation that delivers the goods usually (in a small data set if looking year by year) but not always. Taking the long view, it might be a more reliable driver of returns to expect bear markets to create a significant loss and make up for it with stronger expected returns in bull markets. The max drawdown is 67% for the 60 UPRO / 40 TMF year-to-year spreadsheet backtest.

EfficientInvestor
Posts: 155
Joined: Thu Nov 01, 2018 7:02 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by EfficientInvestor » Fri Jul 12, 2019 5:53 am

MotoTrojan wrote:
Thu Jul 11, 2019 11:08 pm
EfficientInvestor wrote:
Thu Jul 11, 2019 10:22 pm
MotoTrojan wrote:
Thu Jul 11, 2019 9:54 pm
Dockpp wrote:
Thu Jul 11, 2019 9:48 pm

Where do dividend fit in as the future do not pay any?
They don't. But the savings on ER, borrowing rate, and volatility drag could come out ahead. I say could because daily rebalancing (volatility drag) can also create outsized returns (>3x the base asset).
Dividends are accounted for in the pricing of the futures contact. The dividends that are expected over the life of the contract are subtracted from the spot price of the underlying. In other words...you still get the benefit of the dividend, but the benefit is received up front via the discounted purchase price. If this weren’t the case, an arbitrage opportunity would exist in which you could go long the underlying, short the future, and pocket the dividend with zero risk. See the “Forward Price Formula” at the wiki page below.

https://en.m.wikipedia.org/wiki/Forward ... prov=sfti1
Thanks! I must be thinking about LEAPS?
LEAPS also account for dividends. Whenever dividends are paid, the underlying stock price drops by a corresponding amount. Options pricing takes this drop into account. Similar to futures, you receive the benefit of the dividend at the time of purchase in the form of a discount to the price. See article below.

https://www.investopedia.com/articles/a ... prices.asp

jaj2276
Posts: 446
Joined: Sat Apr 16, 2011 5:13 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by jaj2276 » Fri Jul 12, 2019 6:24 am

https://www.bloomberg.com/news/articles ... markets-vp

From the article.
Risk-parity funds are off to their best start since at least 2004 after ramping up exposure to government debt and levering up, while trend followers in interest rates just notched the strongest half-year in nearly three decades.

dspencer
Posts: 176
Joined: Wed Jul 06, 2016 11:29 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by dspencer » Fri Jul 12, 2019 8:07 am

Dr. Long wrote:
Thu Jul 11, 2019 9:06 pm
I really like M1. I currently have 20% of my total portfolio with them - and a quarter of that (5%) devoted to this delightful strategy (experiment?)!
I started the transfer of a 3-fund portfolio at Vanguard to M1 on Jun 10th. It's now July 12th and my money is gone from Vanguard and yet not available at M1. 20% of the balance shows up at M1 but it's not invested and not available to invest. The rest is simply MIA. I have emailed M1 twice and the first time the response was basically "Sorry, there was a slight delay in it being submitted, we'll credit your account $25". That was 2 weeks ago. The next time the response was like an automated message saying my account was paused and to let them know whether I wanted it to resume automatically investing. I did that, but nothing has happened.

So long story short, my experience with M1 so far is awful.

jaj2276
Posts: 446
Joined: Sat Apr 16, 2011 5:13 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by jaj2276 » Fri Jul 12, 2019 8:21 am

dspencer wrote:
Fri Jul 12, 2019 8:07 am
Dr. Long wrote:
Thu Jul 11, 2019 9:06 pm
I really like M1. I currently have 20% of my total portfolio with them - and a quarter of that (5%) devoted to this delightful strategy (experiment?)!
I started the transfer of a 3-fund portfolio at Vanguard to M1 on Jun 10th. It's now July 12th and my money is gone from Vanguard and yet not available at M1. 20% of the balance shows up at M1 but it's not invested and not available to invest. The rest is simply MIA. I have emailed M1 twice and the first time the response was basically "Sorry, there was a slight delay in it being submitted, we'll credit your account $25". That was 2 weeks ago. The next time the response was like an automated message saying my account was paused and to let them know whether I wanted it to resume automatically investing. I did that, but nothing has happened.

So long story short, my experience with M1 so far is awful.
Must be slammed with all the new transfer requests coming in from the people reading this thread!

rascott
Posts: 275
Joined: Wed Apr 15, 2015 10:53 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by rascott » Fri Jul 12, 2019 8:29 am

dspencer wrote:
Fri Jul 12, 2019 8:07 am
Dr. Long wrote:
Thu Jul 11, 2019 9:06 pm
I really like M1. I currently have 20% of my total portfolio with them - and a quarter of that (5%) devoted to this delightful strategy (experiment?)!
I started the transfer of a 3-fund portfolio at Vanguard to M1 on Jun 10th. It's now July 12th and my money is gone from Vanguard and yet not available at M1. 20% of the balance shows up at M1 but it's not invested and not available to invest. The rest is simply MIA. I have emailed M1 twice and the first time the response was basically "Sorry, there was a slight delay in it being submitted, we'll credit your account $25". That was 2 weeks ago. The next time the response was like an automated message saying my account was paused and to let them know whether I wanted it to resume automatically investing. I did that, but nothing has happened.

So long story short, my experience with M1 so far is awful.
That's terrible. I'd be calling/emailing them daily.

Also blast them on the M1 sub on reddit. That should get some attention. An account transfer should take a matter of days.

User avatar
hdas
Posts: 972
Joined: Thu Jun 11, 2015 8:24 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by hdas » Fri Jul 12, 2019 9:04 am

dspencer wrote:
Fri Jul 12, 2019 8:07 am
Dr. Long wrote:
Thu Jul 11, 2019 9:06 pm
I really like M1. I currently have 20% of my total portfolio with them - and a quarter of that (5%) devoted to this delightful strategy (experiment?)!
I started the transfer of a 3-fund portfolio at Vanguard to M1 on Jun 10th. It's now July 12th and my money is gone from Vanguard and yet not available at M1. 20% of the balance shows up at M1 but it's not invested and not available to invest. The rest is simply MIA. I have emailed M1 twice and the first time the response was basically "Sorry, there was a slight delay in it being submitted, we'll credit your account $25". That was 2 weeks ago. The next time the response was like an automated message saying my account was paused and to let them know whether I wanted it to resume automatically investing. I did that, but nothing has happened.

So long story short, my experience with M1 so far is awful.
Just wait until the going gets tough and the weak hands in this thread and elsewhere attempt to spew their positions.

Cheers :greedy
"whenever there is a randomized way of doing something, then there is a nonrandomized way that delivers better performance but requires more thought" ET Jaynes

MotoTrojan
Posts: 5011
Joined: Wed Feb 01, 2017 8:39 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MotoTrojan » Fri Jul 12, 2019 9:44 am

MoneyMarathon wrote:
Fri Jul 12, 2019 2:45 am
Others have made similar comments, but it's worth noting again how different the 1987-2018 data in Portfolio Visualizer is from the 1955-2018 data in Simba's spreadsheet. This is true even if you want to go with just TMF (3x LLT) and UPRO (3x SP500). It seems like the 60 TMF / 40 UPRO balance came from doing portfolio optimization: risk parity and maximizing Sharpe each give about this ratio... on the 1987-2018 data.

Long bonds did well enough during this period that it hardly mattered much (especially without knowing the precise results beforehand, which nobody could know...) whether you were 40/60 favoring bonds, 50/50, or 60/40 favoring stocks. They all had 16-17% CAGR in this time period.

Image

The story's very different in the 1955-2018 data. Then it very much matters for how many millions you end up with.

Image

With CAGR of 11.65%, 12.63%, and 13.25% respectively... where more stocks (up to 60% anyway) meant higher returns.

Standard deviation of 31.24%, 32.48%, and 34.96%. Sharpe of 0.35, 0.38, and 0.39 respectively. Increasing Sharpe and higher CAGR.

You'd have to give up on the idea that the graph will be near-flat in a bear market, but that depends a little on the luck of timing for rebalancing in addition to betting on the negative correlation that delivers the goods usually (in a small data set if looking year by year) but not always. Taking the long view, it might be a more reliable driver of returns to expect bear markets to create a significant loss and make up for it with stronger expected returns in bull markets. The max drawdown is 67% for the 60 UPRO / 40 TMF year-to-year spreadsheet backtest.
This is why I’m so curious how the monthly resetting allocation would’ve done 1955-1987.

Texanbybirth
Posts: 1138
Joined: Tue Apr 14, 2015 12:07 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by Texanbybirth » Fri Jul 12, 2019 10:23 am

I've been following this thread with much interest the last few months. I've decided to take a small % of our NW and implement the strategy, but at 50/50. (I'm not sure if that's been talked about, but does the weighting affect the performance significantly?)

However, I tried at Vanguard and got an error message that we can't buy leveraged products. I spoke with Fidelity (where we have an old IRA) and they said it would be fine. I see M1 Finance mentioned places, but I'd really rather go with a bigger house (especially on the small chance this actually turns into some serious cash in 20 years), so does anyone see a problem with Fidelity?

(Ideally doing this in a Roth IRA is the best choice. I'm currently looking to get one open so we can easily designate that as this test account.)
"Knowledge and innocence are both excellent things, and they are both very funny. But it is right that knowledge should be the servant and innocence the master." - GK Chesterton

JBeck
Posts: 67
Joined: Fri Apr 15, 2016 4:54 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by JBeck » Fri Jul 12, 2019 10:27 am

Texanbybirth wrote:
Fri Jul 12, 2019 10:23 am
I've decided to take a small % of our NW and implement the strategy, but at 50/50.
What made you go with 50/50 as opposed to the recommended 40/60 or even 60/40?

Texanbybirth
Posts: 1138
Joined: Tue Apr 14, 2015 12:07 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by Texanbybirth » Fri Jul 12, 2019 10:32 am

JBeck wrote:
Fri Jul 12, 2019 10:27 am
Texanbybirth wrote:
Fri Jul 12, 2019 10:23 am
I've decided to take a small % of our NW and implement the strategy, but at 50/50.
What made you go with 50/50 as opposed to the recommended 40/60 or even 60/40?
I couldn't decide 60/40 or 40/60 so i split the difference.
"Knowledge and innocence are both excellent things, and they are both very funny. But it is right that knowledge should be the servant and innocence the master." - GK Chesterton

MotoTrojan
Posts: 5011
Joined: Wed Feb 01, 2017 8:39 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MotoTrojan » Fri Jul 12, 2019 10:53 am

Texanbybirth wrote:
Fri Jul 12, 2019 10:23 am
I've been following this thread with much interest the last few months. I've decided to take a small % of our NW and implement the strategy, but at 50/50. (I'm not sure if that's been talked about, but does the weighting affect the performance significantly?)

However, I tried at Vanguard and got an error message that we can't buy leveraged products. I spoke with Fidelity (where we have an old IRA) and they said it would be fine. I see M1 Finance mentioned places, but I'd really rather go with a bigger house (especially on the small chance this actually turns into some serious cash in 20 years), so does anyone see a problem with Fidelity?

(Ideally doing this in a Roth IRA is the best choice. I'm currently looking to get one open so we can easily designate that as this test account.)
If you can’t figure out how to model the differences between these AA’s or at least read this thread enough to find the plots I can’t fathom your desire to jump onboard.

Historically return since 1982 hasn’t varied much with those allocation ranges but drawdown increases dramatically with more equity.

Fidelity is fine but M1 makes the rebalancing seamless.

Texanbybirth
Posts: 1138
Joined: Tue Apr 14, 2015 12:07 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by Texanbybirth » Fri Jul 12, 2019 10:56 am

MotoTrojan wrote:
Fri Jul 12, 2019 10:53 am
Historically return since 1982 hasn’t varied much with those allocation ranges but drawdown increases dramatically with more equity.

Fidelity is fine but M1 makes the rebalancing seamless.
Thank you!
"Knowledge and innocence are both excellent things, and they are both very funny. But it is right that knowledge should be the servant and innocence the master." - GK Chesterton

worthit
Posts: 119
Joined: Tue Jun 19, 2018 2:10 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by worthit » Fri Jul 12, 2019 10:59 am

dspencer, I share your frustrations and agree with you regarding m1. My experience with m1 has been sub-optimal as well to say the least. I had to keep emailing/calling them for nearly 2 weeks before anything would show up in my account. I would never use them as my primary institution. I have less than 2% of my nw invested towards this strategy. I went with them only because of their one click rebalancing option that makes it very easy for the quarterly rebalancing. Not sure what the experience will be when transferring out of m1. I hope it wont be as painful as it was while transfering the account.

MotoTrojan
Posts: 5011
Joined: Wed Feb 01, 2017 8:39 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MotoTrojan » Fri Jul 12, 2019 11:42 am

worthit wrote:
Fri Jul 12, 2019 10:59 am
dspencer, I share your frustrations and agree with you regarding m1. My experience with m1 has been sub-optimal as well to say the least. I had to keep emailing/calling them for nearly 2 weeks before anything would show up in my account. I would never use them as my primary institution. I have less than 2% of my nw invested towards this strategy. I went with them only because of their one click rebalancing option that makes it very easy for the quarterly rebalancing. Not sure what the experience will be when transferring out of m1. I hope it wont be as painful as it was while transfering the account.
Very interesting indeed. I found (and many others based on a recent thread) them to have the BEST customer service of any company I've ever dealt with, not just a brokerage. Super responsive to e-mails, transfer went smoothly, etc... I hope things continue and I will keep an eye on other people's feedback but if this thing grows to $1M I'll have to think a bit harder :mrgreen: .

TNWoods
Posts: 27
Joined: Sun Feb 10, 2019 10:04 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by TNWoods » Fri Jul 12, 2019 3:29 pm

MotoTrojan wrote:
Fri Jul 12, 2019 11:42 am
worthit wrote:
Fri Jul 12, 2019 10:59 am
dspencer, I share your frustrations and agree with you regarding m1. My experience with m1 has been sub-optimal as well to say the least. I had to keep emailing/calling them for nearly 2 weeks before anything would show up in my account. I would never use them as my primary institution. I have less than 2% of my nw invested towards this strategy. I went with them only because of their one click rebalancing option that makes it very easy for the quarterly rebalancing. Not sure what the experience will be when transferring out of m1. I hope it wont be as painful as it was while transfering the account.
Very interesting indeed. I found (and many others based on a recent thread) them to have the BEST customer service of any company I've ever dealt with, not just a brokerage. Super responsive to e-mails, transfer went smoothly, etc... I hope things continue and I will keep an eye on other people's feedback but if this thing grows to $1M I'll have to think a bit harder :mrgreen: .
I also have had a great experience with M1. I created a Roth account, and for some reason I had also created another account, which I didn't mean to. While on the phone explaining it, the rep literally had the other account deactivated before I got another sentence spoken. My account transfer is still in progress, but I have received 4 email updates for each stage of the transfer, and they made it very clear that it could take 7 to 10 days, depending on my current institution's responsiveness.

TNWoods

dave_k
Posts: 331
Joined: Sat Dec 26, 2015 8:25 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by dave_k » Fri Jul 12, 2019 3:37 pm

Texanbybirth wrote:
Fri Jul 12, 2019 10:23 am
However, I tried at Vanguard and got an error message that we can't buy leveraged products. I spoke with Fidelity (where we have an old IRA) and they said it would be fine. I see M1 Finance mentioned places, but I'd really rather go with a bigger house (especially on the small chance this actually turns into some serious cash in 20 years), so does anyone see a problem with Fidelity?
I'm doing this in an IRA and a ROTH at Fidelity. It works fine, but I did have to do something to enable trading leveraged funds and acknowledge the risks (I forget exactly how).

I created a spreadsheet I use after the dividends are paid each quarter that tells me how much of each to buy/sell to reinvest and rebalance.

As a bonus - I have no trading costs for a couple years because I got free trades for transferring money into other brokerage accounts there.

MotoTrojan
Posts: 5011
Joined: Wed Feb 01, 2017 8:39 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MotoTrojan » Fri Jul 12, 2019 4:30 pm

TNWoods wrote:
Fri Jul 12, 2019 3:29 pm
MotoTrojan wrote:
Fri Jul 12, 2019 11:42 am
worthit wrote:
Fri Jul 12, 2019 10:59 am
dspencer, I share your frustrations and agree with you regarding m1. My experience with m1 has been sub-optimal as well to say the least. I had to keep emailing/calling them for nearly 2 weeks before anything would show up in my account. I would never use them as my primary institution. I have less than 2% of my nw invested towards this strategy. I went with them only because of their one click rebalancing option that makes it very easy for the quarterly rebalancing. Not sure what the experience will be when transferring out of m1. I hope it wont be as painful as it was while transfering the account.
Very interesting indeed. I found (and many others based on a recent thread) them to have the BEST customer service of any company I've ever dealt with, not just a brokerage. Super responsive to e-mails, transfer went smoothly, etc... I hope things continue and I will keep an eye on other people's feedback but if this thing grows to $1M I'll have to think a bit harder :mrgreen: .
I also have had a great experience with M1. I created a Roth account, and for some reason I had also created another account, which I didn't mean to. While on the phone explaining it, the rep literally had the other account deactivated before I got another sentence spoken. My account transfer is still in progress, but I have received 4 email updates for each stage of the transfer, and they made it very clear that it could take 7 to 10 days, depending on my current institution's responsiveness.

TNWoods
I also accidentally made a brokerage but just left it with $0. Temptation is there to build a pie and have $20/month go to it for fun but this strategy is enough to keep me busy.

User avatar
abuss368
Posts: 14083
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by abuss368 » Fri Jul 12, 2019 7:15 pm

Many years ago in our evil stock picking days, we “invested” in a two and three time hedge etf. Betting the market would fall.

Speculation at its best.
John C. Bogle: "You simply do not need to put your money into 8 different mutual funds!" | | Disclosure: Three Fund Portfolio + U.S. & International REITs

MoneyMarathon
Posts: 243
Joined: Sun Sep 30, 2012 3:38 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MoneyMarathon » Sat Jul 13, 2019 6:27 am

I have been looking at ways of making a more robust portfolio that includes leveraged ETFs. One of the most reliable way to get long-run returns and a very Bogle-ish idea is to reduce the "drag" from things like expenses. This of course includes the expense ratio and trading costs (e.g. bid/ask spread). It also includes any less-transparent costs. Rebalancing quarterly on an ETF that uses daily-reset leverage can introduce some costs, although they're hard to calculate because it fluctuates and can even be positive (with a chain of increasing daily returns). The simplest way to get a rough estimate might be to look at past results.

I decided to try doing that by comparing the efficiency of different funds to provide 1x or less exposure (specifically 0.9, with the rest in cash). This makes the 2x and 3x funds easier to compare to each other and to a normal ETF. Knowing which leveraged funds are less efficient at providing exposure to the underlying could be one way to know, if you want less than 3x leverage applied to the total portfolio, which normal ETFs are more efficient in a mix (because they're substituting for less-efficient leveraged products).

In each case the underlying product was held at 90% of the portfolio, the 2x product at 45%, and the 3x product at 30% with quarterly rebalancing against cash. I did a comparison for the S&P 500 funds (SPY at 1x / SSO at 2x / UPRO at 3x), for intermediate term 7-10 year bonds (IEF at 1x / UST at 2x / TYD at 3x), and for long-term 20+ year bonds (TLT at 1x / UBT at 2x / TMF at 3x). The time period starts with the youngest ETF in each group (which would be July 2006 for SSO, using the UPROSIM data).

90% SPY came in at 7.71% CAGR, 13.02% std. dev. (0.56 Sharpe... or 1.11 since July 2009).
45% SSO came in at 6.64% CAGR, 13.1% std. dev. (0.48 Sharpe... or 1.04 since July 2009).
30% UPROSIM came in at 6.56% CAGR, 13.08% std. dev. (0.48 Sharpe... or 1.03 since July 2009, with UPRO itself).

90% IEF came in at 3.92% CAGR, 5.03% std. dev. (0.7 Sharpe).
45% UST came in at 3.62% CAGR, 5.05% std. dev. (0.64 Sharpe).
30% TYD came in at 3.8% CAGR, 5.52% std. dev. (0.62 Sharpe).

90% TLT came in at 6.47% CAGR, 11.32% std. dev. (0.57 Sharpe).
45% UBT came in at 6.3% CAGR, 11.74% std. dev. (0.54 Sharpe).
30% TMF came in at 6.17% CAGR, 12.12% std. dev. (0.52 Sharpe).

In each case, using no leverage gave you the best Sharpe, and the biggest step down in Sharpe came from going from no leverage to 2x leverage. There was almost no difference in Sharpe between SSO and UPRO (both leveraged SP500 ETFs from ProShares). Indeed the UPRO portfolio (with more cash) showed less standard deviation than the SSO portfolio, interestingly enough. And there was only a small difference in Sharpe between UST and TYD (the latter being from Direxion, not ProShares) and between UBT and TMF (the latter, again, from Direxion). For each of the 3x bond funds, there was an uptick in standard deviation compared to the 2x bond portfolios, and this could be related to an increase in tracking error, already noted by siamond when trying to model TMF and TYD.

I noticed that TYD in particular seemed to improve its tracking error around 2015, moving to have 0.98 or higher correlation just like its 2x siblings from Proshares, so I figure I could run another test on the latest data set, which might better represent the amount of tracking error to expect from the Direxion 3x leveraged bond funds in the future. Sure enough, the standard deviation discrepancy goes down when Direxion gets better at tracking in a backtest starting Jan 2015.

90% TLT came in at 3.51% CAGR, 10.34% std. dev. (0.3 Sharpe).
45% UBT came in at 2.98% CAGR, 10.44% std. dev. (0.25 Sharpe).
30% TMF came in at 2.96% CAGR, 10.43% std. dev. (0.25 Sharpe).

90% IEF came in at 2.62% CAGR, 4.76% std. dev. (0.38 Sharpe).
45% UST came in at 2.22% CAGR, 4.76% std. dev. (0.3 Sharpe).
30% TYD came in at 2.55% CAGR, 4.85% std. dev. (0.36 Sharpe).

I'm not sure why Direxion is getting such high risk-adjusted returns lately with TYD. The most significant downside of TYD compared to UST right now seems to be that it still has a pretty bad bid-ask spread (0.85% one month average). IEF has lower volatility and lower expenses, including less leverage (which is currently relatively cheap but historically has been more expensive).

For the S&P 500 allocation, if you're not looking for 3x leverage, a split of SPY and UPRO seems more efficient than SSO. Suppose you want to get 0.8x exposure with 40% of the portfolio, applying an average 2x leverage. You seem to be better off mixing UPRO and SPY. (Of course, you might be better off yet by using only UPRO and freeing up other space in the portfolio - compared below.)

80% SPY came in at 7.05% CAGR, 11.56% std. dev. (0.57 Sharpe)
40% SSO came in at 6.11% CAGR, 11.63% std. dev. (0.49 Sharpe)
20% UPRO / 20% SPY came in at 6.29% CAGR, 11.59% std. dev. (0.5 Sharpe)

The gap gets closer than it was when TMF stood alone, so it might be fair to say that there is a benefit to pairing it with TLT but that it is still better to use UBT from ProShares if you want only 2x leverage.

80% TLT came in at 5.87% CAGR, 10.08% std. dev. (0.57 Sharpe)
40% UBT came in at 5.73% CAGR, 10.46% std. dev. (0.54 Sharpe)
20% TMF / 20% TLT came in at 5.68% CAGR, 10.62% std. dev. (0.53 Sharpe)

Fortunately, there is an unleveraged ETF that is better from a portfolio-space perspective at providing exposure to long-term treasury duration risk than TLT. Let's see if we can compare the efficiency of EDV to that of TLT and UBT in terms of risk-adjusted return (Sharpe). To do so, I will match the return of EDV to an 90% TLT portfolio and compare. We will also match the return by using an equal split of TMF and EDV or an equal split of UBT and EDV or an equal split of TMF and UBT and compare the resulting Sharpe to the other options.

90% TLT came in at 6.47% CAGR, 11.32% std. dev. (0.57 Sharpe).
45% UBT came in at 6.3% CAGR, 11.74% std. dev. (0.54 Sharpe).
30% TMF came in at 6.17% CAGR, 12.12% std. dev. (0.52 Sharpe).
60% EDV came in at 6.53% CAGR, 11.89% std. dev. (0.55 Sharpe).
26% UBT / 26% EDV came in at 6.46% CAGR, 11.91% std. dev. (0.55 Sharpe).
21% TMF / 21% EDV came in at 6.54% CAGR, 12.61% std. dev. (0.53 Sharpe).
19% TMF / 19% UBT came in at 6.5% CAGR, 12.61% std. dev. (0.53 Sharpe).

All of these ETFs are very highly correlated, so it's generally a tradeoff of more leverage at the cost of some risk-adjusted return. Once again, though, you may be better off by using only TMF (the most volatile ETF, with the most leverage) and freeing up other space to use to increase risk-adjusted returns.

Adding short term treasuries or cash, however, along with using 3x ETFs, would almost always be a mistake! It's much more efficient (better Sharpe) to reduce your leverage directly (effectively, leverage is short cash yield, indirectly paying LIBOR-based interest costs due to futures and swaps in the leveraged ETFs), instead of adding a cash allocation to a highly-leveraged portfolio. Even though the leverage inside of the ETFs is very cost-effective, returns on cash are generally not going to be higher than the cost of leverage. You can easily deleverage by adding the corresponding regular ETF. The 2x ETFs need a smaller proportion of leverage internally and can also help with deleveraging and possibly improve risk-adjusted returns.

For example, suppose you wanted to target 0.6x exposure to TLT and 0.6x exposure to SPY, with 40% cash to keep all else equal. Moving to using 3x ETFs and using the remaining space for short term treasuries, instead of just using 2x or a mix of 3x and regular ETFs to get the exposure wanted, gave lower Sharpe and lower performance too.

30% SSO, 30% UBT, 40% CASHX came in at 12.68% CAGR, 7.86% std. dev. (1.51 Sharpe).
15% UPRO, 15% SPY, 30% UBT, 40% CASHX came in at 12.78% CAGR, 7.91% std. dev. (1.51 Sharpe).
15% UPRO, 15% SPY, 15% TMF, 15% TLT, 40% CASHX came in at 12.76% CAGR, 7.97% std. dev. (1.5 Sharpe).
20% UPRO, 20% TMF, 20% SHY, 40% CASHX came in at 12.71% CAGR, 8.16% std. dev. (1.46 Sharpe).

Incidentally, a big part of why the 1970s are so terrible for the 3x leveraged LTT / SP500 portfolio isn't just the rising rates hammering LTT, and it isn't just the bear markets at the same time for the SP500. It's also that cash became a relatively-good asset class, because it paid reliably and fairly well, mostly compensating for what was lost to inflation (while also not being exposed to term risk). In times like that, it's very expensive to take a short position on cash. None of the leverage has a fixed rate term locked in. If you think of the full 3x portfolio as having a -200% allocation to cash, the -200% cash allocation is doing poorly at the same time that 3x LTT and the SP500 are. With a 1% expense ratio and various other costs too, it's easy to see why a very similar but unleveraged portfolio would do much better in the 70s. So one of the ways to mitigate the risks of inflation in a portfolio using leveraged ETFs, while also improving risk-adjusted returns, is just to reduce the amount of leverage used to something in-between 1x and 3x.

Summing up, TYD has high bid-ask spreads that make it a bit unattractive for quarterly rebalancing, in addition to a history of tracking error and a recent history of unexplained outperformance (active-ish management?). Unleveraged intermediate-term bonds could be relatively attractive to avoid those drawbacks and also deleverage a portfolio overall. TMF is pretty good at what it does, which is to provide big volatility from term risk even if a small part of the portfolio. In recent years it has been as efficient as UBT. But adding EDV or TLT improves Sharpe a little.

I'd still like to keep looking into other ways to reduce risk in a leveraged ETF portfolio in a 1970s backtest, other than moving to more intermediate duration in bonds, moving to a have a larger ratio of equity to bonds, or deleveraging (all of which help!). Gold might make sense as a diversifier...

MoneyMarathon
Posts: 243
Joined: Sun Sep 30, 2012 3:38 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MoneyMarathon » Sat Jul 13, 2019 2:14 pm

Looking over my notes again, this is an executive summary for long-term investing in leveraged ETFs:
  • Rebalance frequently (e.g. quarterly).
  • Use both stocks and bonds (especially intermediate and long-term treasuries, which tend to hold value when stocks drop).
  • Avoid cash, avoid short-term treasuries. Deleverage instead to improve risk-adjusted performance.
  • Avoid 2x ETFs. Deleverage by using 3x and regular ETFs instead for better risk-adjusted performance.
  • Avoid TYD and UST (leveraged intermediate). UST is only 2x and TYD has issues (thinly traded, tracking error).
  • Consider IEF (7-10 year) or VGIT (5-10 year) for intermediate-term treasuries.
  • Consider EDV (zero-coupon 20+ year STRIPS), TLT (20+ year), or VGLT (10-25 year) alongside TMF (3x 20+ year).
  • Consider SPY or VOO (SP500) or other equity ETFs alongside UPRO (3x SP500).
Other unleveraged ETFs and bond funds also work. Other asset classes, beyond US treasuries and the S&P 500, could also be mixed in. But avoid leveraged ETFs for international stock (based on siamond's notes looking at options such as DZK).

NMBob
Posts: 190
Joined: Thu Apr 23, 2015 8:13 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by NMBob » Sat Jul 13, 2019 2:28 pm

thanks to all for the info.

fyi, if not mentioned before, a 27 page long analysis of leveraged portfolios by some university professors

https://poseidon01.ssrn.com/delivery.ph ... 92&EXT=pdf

MoneyMarathon
Posts: 243
Joined: Sun Sep 30, 2012 3:38 am

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MoneyMarathon » Sat Jul 13, 2019 4:28 pm

I'd like to take a quick look at the leveraged gold ETFs and ETNs. I will use IAU as the no-leverage reference ETF since it very slightly outperformed (with a lower expense ratio than GLD) other old 1x gold options such as GLD and SGOL in historical data. The four largest names in leveraged-bull gold are NUGT, UGL, UGLD, and DGP. Only DGP is rebalanced monthly, the rest daily.

IAU - 1x gold, 0.25% expense ratio, 0.08% bid/ask spread, 0.45 tracking error, 15.32% standard deviation, iShares.
UGL - 2x gold, 0.95% expense ratio, 0.04% bid/ask spread, 0.92 tracking error, 30.57% standard deviation, Proshares.
DGP - 2x gold, 0.75% expense ratio, 0.55% bid/ask spread, rebalanced monthly, 31.24% standard deviation, Deutsche Bank AG.
UGLD - 3x gold, 1.35% expense ratio, 0.05% bid/ask spread, 0.82 tracking error, 46.1% standard deviation, VelocityShares.
NUGT - 3x gold, 1.23% expense ratio, 0.07% bid/ask spread, 1.11 tracking error, 110.85% standard deviation, Direxion.

This correlation matrix makes NUGT in particular look troubled, with extreme volatility and movement less correlated with the rest.

Image

I think that's good enough reason to screen it out for long-term investing and compare IAU, UGL, DGP, and UGLD. First, let's do a Sharpe comparison with IAU against the two 2x gold funds. These all have data from Jan 2009. We'll target 0.9x gold exposure, rest in cash.

90% IAU -> 4.04% CAGR, 15.13% standard deviation, 0.31 Sharpe.
45% UGL -> 2.85% CAGR, 15.87% standard deviation, 0.23 Sharpe.
30% DGP -> 3.29% CAGR, 16.11% standard deviation, 0.26 Sharpe.

This doesn't take into account trading costs due to the larger bid/ask spread on DGP. How much of a drag that is would depend on how much you need to rebalance in and out of DGP, beyond just buying into it. The initial buy-in trading costs can be amortized over a number of years. Even if you took off 0.2% annual CAGR due to trading costs (probably too high), DGP still comes out ahead of UGL, making it look like the better 2x gold fund. I suppose this makes sense from the get-go, given its lower expense ratio (0.2% lower ER should add to return).

There is another set of data starting from Nov 2011, when UGLD became available. Gold was in a downtrend for most of that period, following the commodities boom of the early 21st century, so here it's just better to show a lower loss.

90% IAU -> -2.36% CAGR, 13.78% standard deviation, -0.14 Sharpe.
45% UGL -> -3.2% CAGR, 13.79% standard deviation, 0.21 Sharpe.
45% DGP -> -3.0% CAGR, 14.16% standard deviation, -0.18 Sharpe.
30% UGLD -> -3.16% CAGR, 14.13% standard deviation, -0.2 Sharpe.

So far DGP looks like the least-bad leveraged gold option. But previously we've found that mixing 3x and regular ETFs often provided better Sharpe overall than using 2x. Which is better in this case?

80% IAU -> -1.96% CAGR, 12.25% standard deviation, -0.14 Sharpe.
40% UGL -> -2.72% CAGR, 12.29% standard deviation, -0.21 Sharpe.
40% DGP -> -2.51% CAGR, 12.46% standard deviation, -0.19 Sharpe.
20% UGLD / 20% IAU -> -2.54% CAGR, 12.62% standard deviation, -0.18 Sharpe.

These results make it look like using just the 2x (DGP) and using a mix of 3x and regular (UGLD / IAU) are very close. However, once slippage due to the bid/ask spread is considered as an additional expense for investing in DGP, then the advantage goes to the mix of 3x and regular for those who want to have leverage in gold.

In short, IAU or another regular ETF delivers the best Sharpe (risk-adjusted return), while mixing IAU with UGLD (3x gold) provides the most efficient way (with the best risk-adjusted return) to leverage into gold with LETFs, up to a maximum of 3x with all UGLD.

MotoTrojan
Posts: 5011
Joined: Wed Feb 01, 2017 8:39 pm

Re: HEDGEFUNDIE's excellent adventure [risk parity strategy using 3x leveraged ETFs]

Post by MotoTrojan » Sat Jul 13, 2019 5:21 pm

Just realized ProShares has a 3x inverse long treasury fund, TTT. Odd they don’t have a bull fund in 3x. I’d swap TMF for it.

Post Reply