HEDGEFUNDIE's excellent adventure Part II: The next journey

[kjm]

PV calculates inverse volatility on the Adapativer Asset Allocation. The 4 symbols you used are plotted here: Link

Looks like I'm reinventing the wheel! Are you running something like this?

[perfectuncertainty]

PV calculates inverse volatility on the Adapativer Asset Allocation. The 4 symbols you used are plotted here: Link

Looks like I'm reinventing the wheel! Are you running something like this?

I have a personalized approach in terms of how much I place at risk, but my asset allocation is based on using this algorithm to allocate my investable amount against the symbols I use. I do not use UGL currently. I use TQQQ, QLD, QQQ, TMF and TLT (all tech and all treasuries but I deleverage it with QQQ and TLT).

[langlands]

PV calculates inverse volatility on the Adapativer Asset Allocation. The 4 symbols you used are plotted here: Link

Looks like I'm reinventing the wheel! Are you running something like this?

I have a personalized approach in terms of how much I place at risk, but my asset allocation is based on using this algorithm to allocate my investable amount against the symbols I use. I do not use UGL currently. I use TQQQ, QLD, QQQ, TMF and TLT (all tech and all treasuries but I deleverage it with QQQ and TLT).

Why do you use QQQ, QLD, and TLT? You can deleverage by just going to cash. Is it for tax reasons?

[perfectuncertainty]

PV calculates inverse volatility on the Adapativer Asset Allocation. The 4 symbols you used are plotted here: Link

Looks like I'm reinventing the wheel! Are you running something like this?

I have a personalized approach in terms of how much I place at risk, but my asset allocation is based on using this algorithm to allocate my investable amount against the symbols I use. I do not use UGL currently. I use TQQQ, QLD, QQQ, TMF and TLT (all tech and all treasuries but I deleverage it with QQQ and TLT).

Why do you use QQQ, QLD, and TLT? You can deleverage by just going to cash. Is it for tax reasons?

Of all the leveraged instruments QLD provides the best risk-adjusted returns from my testing (better than TQQQ and UPRO).

Using TQQQ,QLD,QQQ,TMF and TLT the past 10 year return was over 23%. I'll take it.

[langlands]

PV calculates inverse volatility on the Adapativer Asset Allocation. The 4 symbols you used are plotted here: Link

Looks like I'm reinventing the wheel! Are you running something like this?

I have a personalized approach in terms of how much I place at risk, but my asset allocation is based on using this algorithm to allocate my investable amount against the symbols I use. I do not use UGL currently. I use TQQQ, QLD, QQQ, TMF and TLT (all tech and all treasuries but I deleverage it with QQQ and TLT).

Why do you use QQQ, QLD, and TLT? You can deleverage by just going to cash. Is it for tax reasons?

Of all the leveraged instruments QLD provides the best risk-adjusted returns from my testing (better than TQQQ and UPRO).

Using TQQQ,QLD,QQQ,TMF and TLT the past 10 year return was over 23%. I'll take it.

But QLD, QQQ, and TLT can all be obtained as linear combinations of TQQQ, TMF, and CASHX. In other words, I can replicate any portfolio (ignoring expense ratio) you have with those 3 symbols. Now that I think about it, expense ratio does factor in. Holding cash and TQQQ can aways be improved by including some QQQ. Taking into account tax considerations, QLD might find a role as well by minimizing selling.

[Steve Reading]

Looks like I'm reinventing the wheel! Are you running something like this?

I have a personalized approach in terms of how much I place at risk, but my asset allocation is based on using this algorithm to allocate my investable amount against the symbols I use. I do not use UGL currently. I use TQQQ, QLD, QQQ, TMF and TLT (all tech and all treasuries but I deleverage it with QQQ and TLT).

Why do you use QQQ, QLD, and TLT? You can deleverage by just going to cash. Is it for tax reasons?

Of all the leveraged instruments QLD provides the best risk-adjusted returns from my testing (better than TQQQ and UPRO).

Using TQQQ,QLD,QQQ,TMF and TLT the past 10 year return was over 23%. I'll take it.

But QLD, QQQ, and TLT can all be obtained as linear combinations of TQQQ, TMF, and CASHX. In other words, I can replicate any portfolio (ignoring expense ratio) you have with those 3 symbols. Now that I think about it, expense ratio does factor in. Holding cash and TQQQ can aways be improved by including some QQQ. Taking into account tax considerations, QLD might find a role as well by minimizing selling.

This thread is reaching epic levels of data-mining.

[kjm]

Of all the leveraged instruments QLD provides the best risk-adjusted returns from my testing (better than TQQQ and UPRO).

I've found the same thing. There's just too much drag on TQQQ when things get choppy.

You should really look at adding gold. It's fairly uncorrelated with stocks and treasuries - really brings down the volatility from what I've seen.

How often are you adjusting your allocations? You doing this in a regular margin account or in a tax-advantaged account? I have a couple models that backtest pretty well. I'm now looking at how to practically implement them. I ran one in my self-directed 401(k) and triggered some good faith violations... whoops.

[perfectuncertainty]

Of all the leveraged instruments QLD provides the best risk-adjusted returns from my testing (better than TQQQ and UPRO).

I've found the same thing. There's just too much drag on TQQQ when things get choppy.

You should really look at adding gold. It's fairly uncorrelated with stocks and treasuries - really brings down the volatility from what I've seen.

How often are you adjusting your allocations? You doing this in a regular margin account or in a tax-advantaged account? I have a couple models that backtest pretty well. I'm now looking at how to practically implement them. I ran one in my self-directed 401(k) and triggered some good faith violations... whoops.

I'll need to pull out the DD I did on the correlations but QLD is the best bang for the buck compared to UPRO and TQQQ. The risk-adjusted returns show that too. Simply run a HFEA 55/45 PV using QLD/TMF versus UPRO/TMF and it jumps off the page.

[Mickelous]

Of all the leveraged instruments QLD provides the best risk-adjusted returns from my testing (better than TQQQ and UPRO).

I've found the same thing. There's just too much drag on TQQQ when things get choppy.

You should really look at adding gold. It's fairly uncorrelated with stocks and treasuries - really brings down the volatility from what I've seen.

How often are you adjusting your allocations? You doing this in a regular margin account or in a tax-advantaged account? I have a couple models that backtest pretty well. I'm now looking at how to practically implement them. I ran one in my self-directed 401(k) and triggered some good faith violations... whoops.

I'll need to pull out the DD I did on the correlations but QLD is the best bang for the buck compared to UPRO and TQQQ. The risk-adjusted returns show that too. Simply run a HFEA 55/45 PV using QLD/TMF versus UPRO/TMF and it jumps off the page.

Run a TQQQ/TMF Sim as well.

[kjm]

I'll need to pull out the DD I did on the correlations but QLD is the best bang for the buck compared to UPRO and TQQQ.

Hmm... yeah, I'm seeing that. My only concern is the Nasdaq is disproportionately overweight with mega-cap tech, which has significantly outperformed over the past decade. Who's to say that will continue? That style factor could mean-revert and SPY along with LETF variants could outperform, which is why I have both exposures. Man, predicting the future is HARD.

[perfectuncertainty]

Of all the leveraged instruments QLD provides the best risk-adjusted returns from my testing (better than TQQQ and UPRO).

I've found the same thing. There's just too much drag on TQQQ when things get choppy.

You should really look at adding gold. It's fairly uncorrelated with stocks and treasuries - really brings down the volatility from what I've seen.

How often are you adjusting your allocations? You doing this in a regular margin account or in a tax-advantaged account? I have a couple models that backtest pretty well. I'm now looking at how to practically implement them. I ran one in my self-directed 401(k) and triggered some good faith violations... whoops.

I'll need to pull out the DD I did on the correlations but QLD is the best bang for the buck compared to UPRO and TQQQ. The risk-adjusted returns show that too. Simply run a HFEA 55/45 PV using QLD/TMF versus UPRO/TMF and it jumps off the page.

Run a TQQQ/TMF Sim as well.

OK here is what I look at using PV QQQ LINK

BASE QQQ

Code: Select all

Ticker	Leverage	Annualized Return	Return Versus Base	Return versus Leverage
QQQ		1		20.91%		
QLD		2		38.18%			182.592%			91.30%
TQQQ		3		52.98%			253.372%			84.46%

QLD Return versus Leverage (supposed) is superior for less risk.

Same for the S&P: PV SPY LINK

BASE SPY

Code: Select all

Ticker	Leverage	Annualized Return	Return Versus Base	Return versus Leverage
SPY		1		13.82%		
SSO		2		23.31%			168.669%		84.33%
UPRO		3		30.31%			219.320%		73.11%

SSO never comes up but look at the numbers.

Interested in what others see and think.

[coingaroo]

Very interesting analysis! I did some tests to see whether 50% QQQ, 50% TQQQ (net exposure: 2x, same as QLD) would outperform QLD. It actually didn't. I guess vol drag of TQQ outweighs the ER increase. I also believe TQQQ pays higher borrowing costs (given the higher risks involved to lenders) than QLD.

Some thoughts on tech/QQQ:

* QQQ has a higher volatility and std dev than SPY; 23.64% versus 15.04%. Returns over the ETF period are comparable, but I attribute that more to poor timing with the bursting of the dotcom bubble.

* I believe that the market rewards each unit of risk similarly. As an example, the sharpe ratios for stocks, gold, bonds, and even bitcoin are remarkably similar.

* Thus, I believe QQQ will deliver higher returns over the long run, simply because it has more risk. Of course, I also expect it to go down and have steeper drawdowns than SPY. (It hasn't happened in COVID; but you know; COVID and technology and all that).

Some thoughts on deleveraging with QLD/SSO: Personally I am investing in this adventure because I want a portfolio with maximized risk taken for maximum long run return. I'm not trying to get the best risk-adjusted return: if I did there would be far better options than using leveraged ETF funds! That's another reason why I'm TQQQ/TMF 60:40.

---

My biggest concern right now is TMF's yields combined with absurdly high financing costs. I am waiting for the annual report of TMF, coming around October 31st. I want to see just exactly how much interest they are paying on swaps.

The April report said that they were paying apparently 1.2% in interest to borrow for TMF; which is absolutely ludicrous when the LIBOR rate was 0.37% that day. Keep in mind that TMF borrows >250% of its value, only about 50% are in direct long term treasuries in $TLT (which are the collateral).

If you are in UPRO/TQQQ; you don't have to worry. Their borrowing costs were far more reasonable, at like 0.8% or so on the same day.

Furthermore, TMF's swaps are a leverage on the $TLT ETF fund; with a ER of 0.15%. I do not believe this is reported in the TMF SEC ER because they can technically say that it's not an acquired fund expense, they're just buying swaps on the TLT index and it just turns out the TLT index is deflated by TLT ERs.

Looking at the fund fact sheet, they say acquired fund expenses are 0.13%, which is clearly too low for 0.15% x 3 (due to the 3x leverage). So, it seems pretty clear that they are only counting the direct TLT holdings' acquired expenses; and the TLT ER deflator is an additional (undisclosed) drag on returns.

Anyways, let's look at all the fees you are paying with TMF:

• 0.15% * 2.5 in the form of TLT ER tracking drag on swaps = 0.375%
  1.2% * 2.5 in the form of swap financing premiums = 3%
  1.05% in the form of TMF ER = 1.05%

Total real expense ratio of TMF = 4.425% + volatility drag

I don't care about historical performance, when 30 year treasuries were paying 3.45% (Jan 2019 ), 10.35% - (4.425% + libor adjustment) - vol drag = positive.

But now with a yield to maturity of 1.34%, 4.02% - 4.425% - vol drag, it would be an absolute no go. It is an asset with a negative expected value, in nominal terms even, and crazy stdev. You never invest in things with negative expected values: that is called losing money.

I am just hoping the 1.2% swap financing costs (which you pay ~2.5x) was a fluke on that particular day (April 30th) the report was generated, and real financing costs have came down dramatically. Otherwise, I will be replacing TMF.

[Tingting1013]

But now with a yield to maturity of 1.34%, 4.02% - 4.425% - vol drag, it would be an absolute no go. It is an asset with a negative expected value, in nominal terms even, and crazy stdev. You never invest in things with negative expected values: that is called losing money.

You know what else has negative expected value? Insurance. Of all kinds.

[coingaroo]

You know what else has negative expected value? Insurance. Of all kinds.

That doesn't make much sense to me.

1. You never want to actually buy insurance when investing. Like, you wouldn't hold 80% QQQ and 20% SQQQ; it is strictly better to hold 60% QQQ and 40% cash.

2. That example also applies here: if you hold TMF at current ERs and YTMs as "insurance", you'd do better holding less UPRO/TQQQ, and holding (gasp) cash.

3. You can insure against specific events; like insuring against volatility spikes; which will outperform when there are sudden crashes, but it will not protect you when markets decline in a slow and unshocking fashion (i.e. 2008).

4. Modern portfolio theory is about investing in uncorrelated assets for the most risk-efficient return. The diversification improves your sharpe; and the concept of portable beta is about leveraging the optimal (most risk-efficient) allocation to your preferred risk level; to get a higher sharpe than a constrained (unleveraged) portfolio where you must pick a point on the efficient frontier curve.

I would not consider "buying a product that has ~2-5% nominal EV, ~4-7% real EV, ~36% stdev" as within the constructs of portable beta/MPT.

I would consider that to be a strategy/execution mistake; in the same category as holding 80% QQQ and 20% SQQQ, or holding 60% UPRO and putting 40% in Roulette every day in your local casino because its returns are uncorrelated.

And finally, insurance isn't an investment unless you are the one offering insurance. Insurance is protection that you pay a premium for. Colloquially, people say insurance is an 'investment' the same way buying a Tesla for yourself is an 'investment'.

[Tingting1013]

You know what else has negative expected value? Insurance. Of all kinds.

That doesn't make much sense to me.

1. You never want to actually buy insurance when investing. Like, you wouldn't hold 80% QQQ and 20% SQQQ; it is strictly better to hold 60% QQQ and 40% cash.

2. That example also applies here: if you hold TMF at current ERs and YTMs as "insurance", you'd do better holding less UPRO/TQQQ, and holding more cash.

3. You can insure against specific events; like insuring against volatility spikes; which will outperform when there are sudden crashes, but it will not protect you when markets decline in a slow and unshocking fashion (i.e. 2008).

4. Modern portfolio theory is about investing in uncorrelated assets for the most risk-efficient return. The diversification improves your sharpe; and the concept of portable beta is about leveraging the optimal (most risk-efficient) allocation to your preferred risk level; to get a higher sharpe than a constrained (unleveraged) portfolio where you must pick a point on the efficient frontier curve.

I would not consider "buying a product that has ~2-5% nominal EV, ~4-7% real EV, ~36% stdev" as within the constructs of portable beta/MPT.

I would consider that to be a strategy/execution mistake; in the same category as holding 80% QQQ and 20% SQQQ, or holding 60% UPRO and putting 40% in Roulette every day in your local casino because its returns are uncorrelated.

TMF has proven to be negatively correlated to stock crashes. The insurance pays out and then you rebalance. That’s what generates the alpha in this strategy. You can do the same with cash but there is no payout, and so the rebalancing effect is lessened.

Run some back tests of periods when interest rates stayed flat and see how much rebalancing with TMF beats cash.

[coingaroo]

TMF has proven to be negatively correlated to stock crashes. The insurance pays out and then you rebalance. That’s what generates the alpha in this strategy. You can do the same with cash but the effect is lessened.

This mechanism you describe can be obtained strictly cheaper with a short on the index you're trying to invest in, e.g. 60% QQQ, 40% SQQQ.

The only reason why bonds (and TMF) is special is because it has positive expected value; whereas SQQQ has negative expected value (due to fund fees).

Run some back tests of periods when interest rates stayed flat and see how much rebalancing with TMF beats cash.

No problem, I'll do a backtest from Jan 2017 to Dec 2018. Rates stayed relatively flat, with TMF ending where it started (no change). I'll even do quarterly re-balancing which favors UPRO/TMF.

I will then adjust this backtest for current interest rates, because a bond's expected return is its yield to maturity. (You can search for more information; I won't explain it here). See betterbuyandhold.com for more reading on why this is important and mathematically correct (if you choose to not believe it; you're just deluding yourself).


Historical returns at 2.79% starting yield (source: FRED, DGS20)

Code: Select all

				Returns		Stdev		Sharpe(RF=0)
60% UPRO / 40% TMF		13.22%		20.93%		0.63
60% UPRO / 40% CASH		10.86%		20.49%		0.53

Now-casted returns at 1.22% starting yield (source: FRED, DGS20)

Code: Select all

				Returns		Stdev		Sharpe(RF=0)
60% UPRO / 40% TMF		9.56%		20.93%		0.46
60% UPRO / 40% CASH		10.86%		20.49%		0.53

Not so good, does it look? Of course, you'd probably be holding your cash in a HISA, or maybe T-Bills, or maybe even intermediate duration bonds; so cash returns are understated here.

Just to be explicitly clear: I'm not saying TMF is bad because [ rising rates | historically low yields | whatever]. I'm saying TMF is bad because low yields combined with abnormally high swap financing ERs have pushed the expected value to be negative. Whatever anticorrelation properties you can get from it, you can get it cheaper (and better). And this is before the whole "how much can yields go down", which is another can of worms (but it does not help TMF).

I would love for anyone to be able to challenge my numbers here, I really do (I want TMF to work. I'm still invested of it as of posting). But I'd prefer for you to attack it with mathematics; not "it worked before when the EV was positive".

[Tingting1013]

TMF has proven to be negatively correlated to stock crashes. The insurance pays out and then you rebalance. That’s what generates the alpha in this strategy. You can do the same with cash but the effect is lessened.

This mechanism you describe can be obtained strictly cheaper with a short on the index you're trying to invest in, e.g. 60% QQQ, 40% SQQQ.

The only reason why bonds (and TMF) is special is because it has positive expected value; whereas SQQQ has negative expected value (due to fund fees).

Run some back tests of periods when interest rates stayed flat and see how much rebalancing with TMF beats cash.

No problem, I'll do a backtest from Jan 2017 to Dec 2018. Rates stayed relatively flat, with TMF ending where it started (no change). I'll even do quarterly re-balancing which favors UPRO/TMF.

I will then adjust this backtest for current interest rates, because a bond's expected return is its yield to maturity. (You can search for more information; I won't explain it here). See betterbuyandhold.com for more reading on why this is important and mathematically correct (if you choose to not believe it; you're just deluding yourself).


Historical returns at 2.79% starting yield (source: FRED, DGS20)

Code: Select all

				Returns		Stdev		Sharpe(RF=0)
60% UPRO / 40% TMF		13.22%		20.93%		0.63
60% UPRO / 40% CASH		10.86%		20.49%		0.53

Now-casted returns at 1.22% starting yield (source: FRED, DGS20)

Code: Select all

				Returns		Stdev		Sharpe(RF=0)
60% UPRO / 40% TMF		9.56%		20.93%		0.46
60% UPRO / 40% CASH		10.86%		20.49%		0.53

Not so good, does it look? Of course, you'd probably be holding your cash in a HISA, or maybe T-Bills, or maybe even intermediate duration bonds; so cash returns are understated here.

Just to be explicitly clear: I'm not saying TMF is bad because [ rising rates | historically low yields | whatever]. I'm saying TMF is bad because low yields combined with abnormally high swap financing ERs have pushed the expected value to be negative in nominal terms.

You picked a period without any stock crashes and recoveries, of course buying insurance would look bad.

Try August 2012 through July 2019.

[coingaroo]

You picked a period without any stock crashes and recoveries, of course buying insurance would look bad.

Try August 2012 through July 2019.

OK, august 2012 to July 2019. Starting yield 2.55%. Current yield 1.22%.

Nowcasted returns

Code: Select all

				Returns		Stdev		Sharpe(RF=0)
60% UPRO / 40% TMF		21.35%		20.93%		1.02
60% UPRO / 40% CASH		22.48%		20.34%		1.10			

See, again cash strictly better. The edge of TMF came from its historically positive expected value. That edge is now verifiably gone.

Let me add a third one, instead of holding cash, you use more efficient ETFs and achieve the same 2x leverage:

Code: Select all

				Returns		Stdev		Sharpe(RF=0)
60% UPRO / 40% TMF		21.35%		20.93%		1.02
60% UPRO / 40% CASH		22.48%		20.34%		1.10		
80% SSO / 10% UPRO		26.25%		23.57%		1.11

Wow, not only does that beat the nowcasted return, but it even beats the unadjusted UPRO/TMF at a starting yield of 2.55%.

P.S. If you invest based on what performed best in the past, without adjusting for objective changes that affect the present and future, you would be in Exxon Mobil and Shell.

[HawkeyePierce]

You can sidestep a lot of the issues with TMF by simply deleveraging to EDV and accepting a slightly smaller UPRO allocation.

[langlands]

You picked a period without any stock crashes and recoveries, of course buying insurance would look bad.

Try August 2012 through July 2019.

OK, august 2012 to July 2019. Starting yield 2.55%. Current yield 1.22%.

Nowcasted returns

Code: Select all

				Returns		Stdev		Sharpe(RF=0)
60% UPRO / 40% TMF		21.35%		20.93%		1.02
60% UPRO / 40% CASH		22.48%		20.34%		1.10			

See, again cash strictly better. The edge of TMF came from its historically positive expected value. That edge is now verifiably gone.

Let me add a third one, instead of holding cash, you use more efficient ETFs and achieve the same 2x leverage:

Code: Select all

				Returns		Stdev		Sharpe(RF=0)
60% UPRO / 40% TMF		21.35%		20.93%		1.02
60% UPRO / 40% CASH		22.48%		20.34%		1.10		
80% SSO / 10% UPRO		26.25%		23.57%		1.11

Wow, not only does that beat the nowcasted return, but it even beats the unadjusted UPRO/TMF at a starting yield of 2.55%.

P.S. If you invest based on what performed best in the past, without adjusting for objective changes that affect the present and future, you would be in Exxon Mobil and Shell.

I just want to point out that 80% + 10% doesn't add up to 100%. Also, if you meant 80% SSO and 20% UPRO, that comes out to 2.2 leverage, which is more than the 1.8 leverage you get from 60% UPRO/40%CASH. So something isn't quite right with that comparison.

Regarding TMF, I hope someone good at digging through this kind of thing (there were a few people earlier on who did a really good job of working through the details of the funding costs) can verify that TMF funding costs are in fact unreasonably high. I seemed to remember at least last year that the story was the reverse: UPRO had high funding costs and TMF had very favorable funding costs. It's possible things have changed. Also, there is a tradeoff between expected value and correlation. If the correlation between two assets is negative enough, a negative expected value can be perfectly acceptable. Just imagine an extreme scenario where stocks return 10%, bonds return -5% and the correlation between them is -0.95. I would definitely load up on a lot of both stocks and bonds (in about equal amount) and pick up a near risk free 5% return.

But overall, yes I share your concerns about TMF. I think there are signs the correlation between bonds and stocks is moderating, decreasing the diversification benefit. The reduced expectation of bond returns also doesn't help. Put very simply, bonds are a much less attractive investment. Some people think stocks are also not a great investment now. Pick your poison, I guess.

[Mickelous]

You can sidestep a lot of the issues with TMF by simply deleveraging to EDV and accepting a slightly smaller UPRO allocation.

Could you summarize how would EDV be more beneficial?

[occambogle]

Could you summarize how would EDV be more beneficial?

I’m guessing because expense ratio is only 0.07% and no volatility decay. That said if one feels TMF isn’t providing enough “insurance” then EDV will be even less so.

[HawkeyePierce]

Could you summarize how would EDV be more beneficial?

I’m guessing because expense ratio is only 0.07% and no volatility decay. That said if one feels TMF isn’t providing enough “insurance” then EDV will be even less so.

There are also no hidden borrowing expenses. WYSISWYG with EDV.

[perfectuncertainty]

Why not add in the base symbols like QQQ/SPY with TLT using AAA Inverse Volatility. The results are still impressive and the risk is lower.

[NMBob]

You can sidestep a lot of the issues with TMF by simply deleveraging to EDV and accepting a slightly smaller UPRO allocation.

Could you summarize how would EDV be more beneficial?

poster mototrojan was a proponent of edv with many posts of information.

https://www.google.com/search?sitesearc ... mototrojan

[Mickelous]

I found 60/40 EDV/TQQQ produced similar results to 55/45 UPRO/TMF, 65/35 EDV/TQQQ produced a smudge lower but had much better drawdowns. There seems to be a nice balance between the two assets.

[ATLGator]

Related question - Are there any brokerages that will allow me to input my asset allocation across these ETFs and do the rebalancing for me automatically (on some schedule)? Preferably without an additional management fee.

[rchmx1]

Related question - Are there any brokerages that will allow me to input my asset allocation across these ETFs and do the rebalancing for me automatically (on some schedule)? Preferably without an additional management fee.

Look into M1. I don't know if it's completely automatic, but people in this thread have mentioned M1 as offering convenient rebalancing features.

[Mickelous]

Related question - Are there any brokerages that will allow me to input my asset allocation across these ETFs and do the rebalancing for me automatically (on some schedule)? Preferably without an additional management fee.

Look into M1. I don't know if it's completely automatic, but people in this thread have mentioned M1 as offering convenient rebalancing features.

I checked for myself and I was unable to find auto rebalancing timeframes, however its simple to rebalance as you just go in the app or website and click the rebalance button.

[cos]

Related question - Are there any brokerages that will allow me to input my asset allocation across these ETFs and do the rebalancing for me automatically (on some schedule)? Preferably without an additional management fee.

Look into M1. I don't know if it's completely automatic, but people in this thread have mentioned M1 as offering convenient rebalancing features.

I checked for myself and I was unable to find auto rebalancing timeframes, however its simple to rebalance as you just go in the app or website and click the rebalance button.

I have no idea if this still works since it was last updated in 2019, but here's an unofficial API library written in Python: https://github.com/etz/m1lib

If you can get that working, you can write your own script to automate the rebalancing process.

[rchmx1]

Related question - Are there any brokerages that will allow me to input my asset allocation across these ETFs and do the rebalancing for me automatically (on some schedule)? Preferably without an additional management fee.

Look into M1. I don't know if it's completely automatic, but people in this thread have mentioned M1 as offering convenient rebalancing features.

I checked for myself and I was unable to find auto rebalancing timeframes, however its simple to rebalance as you just go in the app or website and click the rebalance button.

Ya, that's what I remember reading. M1 won't do it themselves, but once you set your allocations it just requires one click of a button. It seems like that's a close as it gets to auto rebalancing, without paying an advisor to do it for you.

[BuffMaltese]

For those of you using MV or IV methods, is it concerning that by mid month the AAA can be drastically different than what was recommended at the beginning of the month? This month, using qld/tmf with MV we started at 61% qld which shifted to recommending 21% qld. Do any of you make exceptions for sticking with the recommendation at the start of the month if there is a certain percentage change. Am I thinking about this wrong? Seems odd to me to stick with something that shifted so far.

[kjm]

For those of you using MV or IV methods, is it concerning that by mid month the AAA can be drastically different than what was recommended at the beginning of the month? This month, using qld/tmf with MV we started at 61% qld which shifted to recommending 21% qld. Do any of you make exceptions for sticking with the recommendation at the start of the month if there is a certain percentage change. Am I thinking about this wrong? Seems odd to me to stick with something that shifted so far.

There's no rule saying you have to wait a month to rebalance. Returns and risk metrics are actually better (with inverse volatility at least) if you rebalance weekly. If you're in a taxable account, you'll run into the wash rule and lose the ability to deduct your loses. However, that loss may be less than what you make by rebalancing more frequently. Plus, it may provide some peace of mind. If you're in a tax-advantaged account, I don't see any drawback to rebalancing weekly... assuming you're not still paying commission fees somehow.

[d9i]

Hey folks, I'm new here, and I'm still making my way through the thread.

I was curious what everyone's thoughts are on TMF/TLT's continued usefulness as a hedge against downturns in equities, now that interest rates are near 0.

Is there a belief that future Fed activity (increased QE, average inflation targeting) will make bonds more attractive? Are people banking on negative interest rates? I noticed some people looking at gold as an alternative hedge, but it clearly doesn't have the same characteristics that bonds do.

[CanaBogle24]

I have $50k set aside to put into this strategy (which equates to ~10% of current investable assets). I'm planning to use the modified original allocation of 55/45 UPRO/TMF (though I am flirting with splitting the UPRO allocation into 50/50 UPRO/TQQQ).

What I'm wondering is, given current market volatility, if it's worth either holding off a while or to DCA in over some pre-defined period of time. With my normal Boglehead-minded investments, I'm fine just dumping whatever I have in (and have been doing so over the past few quarters). But with this strategy in particular, noting volatility decay, I'm tempted to hold off until there's a bit more social and market stability.

To pre-empt a few questions / challenges...

What constitutes market stability to you?
In particular, either a vaccine, emerging herd immunity or a more entranched way of "living with" an active pandemic, not to mention being on the other side of the impending election

But what if we never hit a time that fits your description of "stable market conditions" or new uncertainty arises (as it always does)?
One option would be to set a timeline to hold myself to this; e.g. if things haven't stabilized by [arbitrary date 6-12 months from now], just take the plunge

But what if the market just keeps climbing between now and the arrival of sweet stability?
Well, then I miss out on that upside, obviously; but is missed upside not just the other side of the risk coin? I'm not trying to "time the market" in the way of trying to wait for some particular dip or market low. I just feel like uncertainty about market direction over the next ~year is at a generational high. This isn't "I have a hunch we're due for a correction, so I'll wait it out". This is more "nobody knows what the hell's going on in the world, maybe I should wait until the waters calm a bit given sideways movement means a decline for this strategy."

I know this sounds like I'm trying to convince myself, but knowing I'm usually not a market timer at all and follow the lump sum over DCA philosophy, I'm more than open to be convinced otherwise here. I'm not worried about the psychological impact of going in now and seeing my holdings decline substantially - I can definitely deal with that and am certain I'll stick it out for the long term (I'm in my mid thirties and plan to hold this for decades). I'm just wondering if it'd be a more pragmatic choice to be patient under current conditions.

Thanks so much for your thoughts!

[z1_gravedigger]

I see nothing wrong with dollar cost averaging over the next 6 - 12 months. This is an opinion, and I don’t think you’ll find a definitive “correct” answer.

Waiting for the “right moment” to jump in sounds like market timing at best, or at worst, indicates your risk tolerance may not be as high as you believe. I mean no disrespect, but if one recession or election is giving you pause, HFEA isn’t the strategy.

[kjm]

Hey folks, I'm new here, and I'm still making my way through the thread.

I was curious what everyone's thoughts are on TMF/TLT's continued usefulness as a hedge against downturns in equities, now that interest rates are near 0.

Is there a belief that future Fed activity (increased QE, average inflation targeting) will make bonds more attractive? Are people banking on negative interest rates? I noticed some people looking at gold as an alternative hedge, but it clearly doesn't have the same characteristics that bonds do.

The Fed is still buying government bonds at a rate of about $20 billion per week. QE isn't going anywhere. And, rates can always go negative. Ultimately, I would say trust the system - whatever that may be. The nice thing about going the inverse volatility route for asset allocation is that the TLT portion of your portfolio will shrink in proportion to risk. I don't think rates will rise until inflation kicks in, which is still years out if you believe the fed. So, at the very least, treasuries should reduce the overall volatility of your portfolio.

Gold has a strong negative correlation to the dollar. If the recession we're in continues longer than expected or deepens, the dollar will rise and gold will likely fall. However, enough fiscal/monetary stimulus could put negative pressure on the dollar. In fact, I think a devalued dollar is what the Fed is after. All this to say you guess is as good as mine.

[Mickelous]

I am really liking this strategy using a high amount of EDV between 57-60 percent and 40-43 percent TQQQ. Only 180% leveraged vs 300% with similar returns to upro/tmf 55/45 in the last bull run. Haven't done long run backtesting with simulated data as I don't have it for the other tickers. Just switched over in my 401k and roth to this compared to upro/tqqq/tmf HFEA that I had before.

[rchmx1]

z1 makes some good points. My feeling is that I don't think the lump sub vs DCA consideration is really that different in this strategy. The future market behavior will determine which of those options was "right," and since we can't know the future there's no useful insight to be drawn to help inform our present decision.

Something that has been pointed out quite frequently in this thread is that the starting point does make a big difference in the outcome of this strategy (also depending upon timeframe). So, if you started this strategy in mid February of this year, when UPRO was trading around $80, you'd be having a much different experience than if you started around mid March, when UPRO was trading in the $20s.

Recognizing the inherent uncertainty of the market, and the fact that leverage enhances that uncertainty, I think the best you can do in terms of trying to insure that you start this strategy at an optimal time is to begin after/during some kind of dip. We've just had a dip. UPRO got up to around $70 a few weeks ago, and is now trading in the low $50s. TQQQ got up to around $180 a few weeks ago, and is now trading around $120. Is that a big enough dip to make you feel comfortable starting now? Do you want to start now with a portion of your in the hopes that there will be another larger dip forthcoming? Questions you'll have to decide for yourself.

[kjm]

What I'm wondering is, given current market volatility, if it's worth either holding off a while or to DCA in over some pre-defined period of time.

DCA in or wait 6 weeks until after the election if you have cold feet. IMO, the biggest risk to markets now is Constitutional/electoral crisis on November 3rd. That would be uncharted territory, and the market hates uncertainty more than anything.

The case for jumping in now would the fact that VIX has been trending down the past few weeks, which could be a signal that the current bout of volatility is coming to an end. Who really knows though!?

[nehawk87]

I have officially rolled some of my Roth to M1 Finance and executed my trades into 55/45 UPRO/TMF today.

I’m into UPRO at $50.03/share.

Also, the M1 trading structure is really interesting. I personally loved just setting up the “pie” and letting M1 execute the trade in the AM. Obviously this will be a breeze for rebalancing too as others have mentioned.

[manlymatt83]

How do you folks allocate UPRO/TMF into your overall asset allocation?

[zie]

How do you folks allocate UPRO/TMF into your overall asset allocation?

Enough to notice if it rocks, not enough that it matters if it all goes away. I.e. not enough that it will affect my retirement noticeably.

There is a lot of risk adopting this strategy, so going all-in would probably not be very wise, however the potential rewards if it does as well as the last decade would be pretty amazing!

[manlymatt83]

How do you folks allocate UPRO/TMF into your overall asset allocation?

Enough to notice if it rocks, not enough that it matters if it all goes away. I.e. not enough that it will affect my retirement noticeably.

There is a lot of risk adopting this strategy, so going all-in would probably not be very wise, however the potential rewards if it does as well as the last decade would be pretty amazing!

So in your “spreadsheet” do you calculate it as a % relative to something else? Like ... 10% of total market holdings? Or is it just something you keep fixed dollar in a separate account and say “meh I’ll see what it is when I retire”? I guess the biggest question is... would you ever rebalance into or out of it?

[stimulacra]

How do you folks allocate UPRO/TMF into your overall asset allocation?

Targeting 5% by 2021. Aside from quarterly rebalancing, I'll let the M1 Finance Roth account just ride till 2037.

It's actually encouraging me to de-slice and de-tilt my non-leveraged portfolio and back closer to a 3-fund portfolio.

[zie]

How do you folks allocate UPRO/TMF into your overall asset allocation?

Enough to notice if it rocks, not enough that it matters if it all goes away. I.e. not enough that it will affect my retirement noticeably.

There is a lot of risk adopting this strategy, so going all-in would probably not be very wise, however the potential rewards if it does as well as the last decade would be pretty amazing!

So in your “spreadsheet” do you calculate it as a % relative to something else? Like ... 10% of total market holdings? Or is it just something you keep fixed dollar in a separate account and say “meh I’ll see what it is when I retire”? I guess the biggest question is... would you ever rebalance into or out of it?

I personally went backwards. I need X dollars to make my retirement goals (say 25x yearly estimated expenses in retirement @ 4% SWR) -- calculated in 2020 dollars(so no inflation included). To make that number from where I am today given 5%/yr growth(i.e. TSM growth minus inflation so it's also in 2020 dollars) I need to invest $X/yr(for example let's say it's $10k/yr). So anything over and above that $10k/year I make I can spend on anything I want. Obviously things like food, shelter etc have to come out, but then I have $X/yr play money. This is the money I'm spending on stuff like this strategy.

I.e. it won't affect my retirement, as that's been calculated out. My food and shelter and whatever else I want to spend has also been calculated out, so whatever is left I can waste on whatever I wish. Right now it's currently buying UPRO/TMF. As time goes on, I will probably decide to spend that extra $x/yr play money on something else, say a new computer someday, whatever. It doesn't really matter what I do with that money, it doesn't affect my retirement plans or my ability to eat and sleep comfortably.

Some people in Bogleheads land calculate 1-10% of their savings money for "play". I personally take the opposite view, I need $X/yr to meet my current and future goals(to include retirement obviously), anything I make above that number I can spend on whatever I wish, and I do.

I would never rebalance out of it, unless I found something else more interesting to do with the money(which I haven't yet). I am currently buying into it every month. It's not a lot of money in real terms yet, but that doesn't matter.

[mistermojorizin]

Has anyone looked at doing something like 25 percent tmf and 75 percent spy?

Strategies involving upro produce great results and all but higher standard deviations. The spy/tmf strategy seems to produce better results and lower standard deviation compared to spy alone.

[am]

Is this strategy still worth doing if the leveraged long bonds have so little upside left given ultra low interest rates?

[NMBob]

Is this strategy still worth doing if the leveraged long bonds have so little upside left given ultra low interest rates?

i suggest you look through the two hef threads. there must be at least 1500-3500 previous posts on that issue, not all in agreement.

[manlymatt83]

How do you folks allocate UPRO/TMF into your overall asset allocation?

Enough to notice if it rocks, not enough that it matters if it all goes away. I.e. not enough that it will affect my retirement noticeably.

There is a lot of risk adopting this strategy, so going all-in would probably not be very wise, however the potential rewards if it does as well as the last decade would be pretty amazing!

So in your “spreadsheet” do you calculate it as a % relative to something else? Like ... 10% of total market holdings? Or is it just something you keep fixed dollar in a separate account and say “meh I’ll see what it is when I retire”? I guess the biggest question is... would you ever rebalance into or out of it?

I personally went backwards. I need X dollars to make my retirement goals (say 25x yearly estimated expenses in retirement @ 4% SWR) -- calculated in 2020 dollars(so no inflation included). To make that number from where I am today given 5%/yr growth(i.e. TSM growth minus inflation so it's also in 2020 dollars) I need to invest $X/yr(for example let's say it's $10k/yr). So anything over and above that $10k/year I make I can spend on anything I want. Obviously things like food, shelter etc have to come out, but then I have $X/yr play money. This is the money I'm spending on stuff like this strategy.

I.e. it won't affect my retirement, as that's been calculated out. My food and shelter and whatever else I want to spend has also been calculated out, so whatever is left I can waste on whatever I wish. Right now it's currently buying UPRO/TMF. As time goes on, I will probably decide to spend that extra $x/yr play money on something else, say a new computer someday, whatever. It doesn't really matter what I do with that money, it doesn't affect my retirement plans or my ability to eat and sleep comfortably.

Some people in Bogleheads land calculate 1-10% of their savings money for "play". I personally take the opposite view, I need $X/yr to meet my current and future goals(to include retirement obviously), anything I make above that number I can spend on whatever I wish, and I do.

I would never rebalance out of it, unless I found something else more interesting to do with the money(which I haven't yet). I am currently buying into it every month. It's not a lot of money in real terms yet, but that doesn't matter.

Super helpful! Thank you!