HEDGEFUNDIE's excellent adventure Part II: The next journey

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
IntolerableOptimism
Posts: 5
Joined: Fri Jul 10, 2020 10:28 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by IntolerableOptimism »

I have done more research on using levered funds for a risk parity strategy, and stumbled across a very math intensive paper on leverage and risk parity strategies.

https://riskcenter.berkeley.edu/wp-cont ... r1_000.pdf
Of note is table 8 on page 23 on the overall returns of the different strategies from 1929 - 2012. From my interpretation of the results, a 60/40 leveraged portfolio outperformed a non-leveraged 60/40 portfolio, however, it did not outperform a non-leveraged 100% stock portfolio.

Based on the very heady math involved, it seems the covariance between leverage and the excess returns of the target portfolio needs to be accounted for during backtesting. I'm not sure if this can be done using portfoliovisualizer or other backtesting software.

The paper above is in response to another research paper on leverage and risk parity:
https://www.q-group.org/wp-content/uplo ... parity.pdf

The main point in the study for me is on slide 50. Taking into account financing costs the probability of the stock market beating a 60/40 risk parity strategy is 52% over the next 20 years and 54% over the next 50 years.

This strategy isn't a free lunch and requires a lot of babysitting. Essentially, a full-time leveraged, risk parity strategy will underperform a non-leveraged 100% stock portfolio in the long term. There will be times when it makes sense to use leverage, but only when future volatility is predicted to be below its average. Moving forward, I am not planning on increasing my leveraged positions. I'm in the process of building a GARCH model to predict future volatility (still researching accuracy of how many days in the future GARCH modeling can forecast with statistical significance).

My final takeaway is that leverage does work over certain periods (variable given current and future volatility). Outside of fees, volatility has a large impact on the use of leverage. Since volatility is easier (nowhere near perfect forecasting, just easier) to forecast than returns, it would make sense to add leverage to a portfolio during time periods that have low volatility and/or forecasted to have lower than average volatility (https://www.sas.upenn.edu/~fdiebold/pap ... wnlees.pdf).
Semantics
Posts: 185
Joined: Tue Mar 10, 2020 1:42 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Semantics »

IntolerableOptimism wrote: Mon Aug 03, 2020 11:16 am I have done more research on using levered funds for a risk parity strategy, and stumbled across a very math intensive paper on leverage and risk parity strategies.

https://riskcenter.berkeley.edu/wp-cont ... r1_000.pdf
Of note is table 8 on page 23 on the overall returns of the different strategies from 1929 - 2012. From my interpretation of the results, a 60/40 leveraged portfolio outperformed a non-leveraged 60/40 portfolio, however, it did not outperform a non-leveraged 100% stock portfolio.

Based on the very heady math involved, it seems the covariance between leverage and the excess returns of the target portfolio needs to be accounted for during backtesting. I'm not sure if this can be done using portfoliovisualizer or other backtesting software.

The paper above is in response to another research paper on leverage and risk parity:
https://www.q-group.org/wp-content/uplo ... parity.pdf

The main point in the study for me is on slide 50. Taking into account financing costs the probability of the stock market beating a 60/40 risk parity strategy is 52% over the next 20 years and 54% over the next 50 years.

This strategy isn't a free lunch and requires a lot of babysitting. Essentially, a full-time leveraged, risk parity strategy will underperform a non-leveraged 100% stock portfolio in the long term. There will be times when it makes sense to use leverage, but only when future volatility is predicted to be below its average. Moving forward, I am not planning on increasing my leveraged positions. I'm in the process of building a GARCH model to predict future volatility (still researching accuracy of how many days in the future GARCH modeling can forecast with statistical significance).

My final takeaway is that leverage does work over certain periods (variable given current and future volatility). Outside of fees, volatility has a large impact on the use of leverage. Since volatility is easier (nowhere near perfect forecasting, just easier) to forecast than returns, it would make sense to add leverage to a portfolio during time periods that have low volatility and/or forecasted to have lower than average volatility (https://www.sas.upenn.edu/~fdiebold/pap ... wnlees.pdf).
Thanks for posting these links, it's interesting to see an academic take on this. I do have one qualm though, which is that the conclusions are still based on a backtest. It's a much more rigorously constructed one than most of what's discussed in this thread, but nonetheless depends on market conditions, and doesn't allow any flexibility to deviate from 55/45 even if it becomes (I doubt most people here are committed to it for eternity). My main takeaway echoes your last one - the approach works over certain time periods.

The chart on page 12 is key. It shows that leverage has done extremely well from 1982-present, and really suffered from 1963-1981, corresponding to the treasury bond market. This has been a major point of discussion around here. I plan to stick to the strategy at a high level, but move out of long term treasuries if it starts to look like inflation is being priced in, shifting some weight into more equities, some into bond alternatives (maybe BTAL), and maybe shorter term treasuries.

Couple other notes about the methodology that I think make the approach look less favorable:
  • They assume trading costs of 1% until 1955 and .5% until 1971, which may have been historically accurate, but there's no reason to think they will be that high in the future.
  • The historical borrowing rate they use is the 3-month T-bill + 60 bps, but leveraged ETFs get much better borrowing costs with swaps (though maybe you can call it a wash after expense ratios).
  • Their leverage fluctuates a lot, and apparently was 3.66 on average, since they aim for a target volatility. That seems a little different than what we are doing here. Need to read the paper more to understand the full implications.
pleonasm
Posts: 30
Joined: Tue Oct 15, 2019 11:59 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by pleonasm »

Volatility Decay means sequence of daily returns via arithmetic exposure. This is exactly how the Swap contracts work. You do not have a true nominal exposure like you do with futures, it's true you have the 3x exposure of your value, but it is entirely arithmetic fueled by swaps.

That being said, again SQQQ short vs TQQQ long is all about where it is on the arithmetic acceleration curve. PV is broken by reverse splits, throw out all the backtests, they are garbage. It only matters what the price is at the moment you enter if TQQQ or SQQQ short is better for total returns or risk.

https://www.splithistory.com/tqqq/
https://www.splithistory.com/sqqq/

There's your true backtest.
SQQQ has returned 2500%
TQQQ has returned 6126%

But again, only the share price matters due to the arithmetic. Thank goodness for swaps that make this easy.

At today's price of TQQQ $124 & SQQQ at $5.6, TQQQ has far more arithmetic acceleration. If volatility is high, acceleration is bad, if it is low, it is good.

It is not a function of the market or how these products behave (internally). It is entirely about price.

That said, I use options to risk adjust instead of using risk parity.
000
Posts: 4119
Joined: Thu Jul 23, 2020 12:04 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by 000 »

Why no junk bonds? Seems less interest rate sensitive and more fundamentally sound.

See: viewtopic.php?f=10&t=321924
Thereum
Posts: 62
Joined: Sun Jun 14, 2020 9:05 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Thereum »

000 wrote: Mon Aug 03, 2020 10:40 pm Why no junk bonds?
Because they are strongly correlated with stocks.

On my short SQQQ/TMV strategy something interesting happened -- I got assigned early. Only this wasn't SQQQ but TECS, which is a very similar ETF (0.97 correlation) but with much lower liquidity and higher short selling fees. I now have a synthetic put option instead of a short call spread. The buyer gifted me the time premium but now I am stuck paying a much higher short selling fee compared to SQQQ. I suppose I underestimated the importance of liquidity in this strategy.

My other thought is that whoever bought this option is likely a sophisticated market maker. Did they exercise the option to get a higher delta because they anticipate an upcoming correction? It's all very confusing to me. Also, why can't I just buy back the shares and sell the exact same option and collect time premium again? To prevent this in the future, I will sell a call with slightly more time premium.

(Edit: On further analysis, it appears the market maker thinks they will win on the bid-ask spread if I try to cover and reenter, and they are right. Lesson learned: Do not sell options with minimal time value in illiquid underlyings.)

@pleonasm

When I backtest options strategies, I get better results for shorting SQQQ/TMV than I do going long on TQQQ/TMF. I don't think I have found any long strategy that beats going short in terms of risk-adjusted and absolute returns. That said, going long TQQQ could return more in a calm bull market, due to the continuous compounding of daily gains. I believe it will struggle if volatility returns, and I think the LTT hedge is especially risky with rates so low.
pleonasm
Posts: 30
Joined: Tue Oct 15, 2019 11:59 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by pleonasm »

Thereum wrote: Tue Aug 04, 2020 12:04 am
000 wrote: Mon Aug 03, 2020 10:40 pm Why no junk bonds?
Because they are strongly correlated with stocks.

@pleonasm

When I backtest options strategies, I get better results for shorting SQQQ/TMV than I do going long on TQQQ/TMF. I don't think I have found any long strategy that beats going short in terms of risk-adjusted and absolute returns. That said, going long TQQQ could return more in a calm bull market, due to the continuous compounding of daily gains. I believe it will struggle if volatility returns, and I think the LTT hedge is especially risky with rates so low.
If you are using PV to backtest, your backtests are broken. PV is not calculating the reverse splits correctly. Again, the difference is arithmetic acceleration. There is a very good reason why no one is short SQQQ and everyone is long TQQQ. Over the past 10 years, the reason is 3600%.
Take these examples
The Q's
Image

The Bonds
Image

Mixed
Image

It's broken. The mix makes no sense and cannot be reconciled to our quasi-rule based financial markets. Because of the daily rebalance, shorting SQQQ when it is a large # in comparison to TQQQ is when it outperforms. PV cannot show you when that is because the data is borked.
Semantics
Posts: 185
Joined: Tue Mar 10, 2020 1:42 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Semantics »

pleonasm wrote: Mon Aug 03, 2020 10:37 pm Volatility Decay means sequence of daily returns via arithmetic exposure. This is exactly how the Swap contracts work. You do not have a true nominal exposure like you do with futures, it's true you have the 3x exposure of your value, but it is entirely arithmetic fueled by swaps.

That being said, again SQQQ short vs TQQQ long is all about where it is on the arithmetic acceleration curve. PV is broken by reverse splits, throw out all the backtests, they are garbage. It only matters what the price is at the moment you enter if TQQQ or SQQQ short is better for total returns or risk.

https://www.splithistory.com/tqqq/
https://www.splithistory.com/sqqq/

There's your true backtest.
SQQQ has returned 2500%
TQQQ has returned 6126%

But again, only the share price matters due to the arithmetic. Thank goodness for swaps that make this easy.

At today's price of TQQQ $124 & SQQQ at $5.6, TQQQ has far more arithmetic acceleration. If volatility is high, acceleration is bad, if it is low, it is good.

It is not a function of the market or how these products behave (internally). It is entirely about price.

That said, I use options to risk adjust instead of using risk parity.
I think you missed a couple of zeros. I get 237800% for SQQQ (~2500x).
Thereum
Posts: 62
Joined: Sun Jun 14, 2020 9:05 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Thereum »

pleonasm,

I understand why you are getting your result. You need to allow monthly rebalancing for the short-only portfolios.

Make a portfolio with -100% SQQQ and 100% CASHX and rebalance monthly. You should see the right results with this.
euphonious
Posts: 32
Joined: Tue Aug 04, 2020 2:43 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by euphonious »

For the timing models on PV, are the "Forward signals" (click on Timing Periods and scroll down) simply based on implied volatility of options a month out, or something else?

Also, for those who are rebalancing based on volatility, what's the reason for using historical volatility instead of forward looking volatility (e.g. VIX for UPRO / VXN for TQQQ, or the IV from their options)? Have there been any backtests done using forward looking volatility?

One last question, for those who use a 1 month volatility window, are you all rebalancing every month? Based on results from Refinements to Hedgefundie's excellent approach:
The volatility window should be tuned to the rebalance frequency. Performance degrades if the volatility window is much shorter than the rebalance frequency.
Does that mean for those who want to rebalance quarterly, the preferred volatility window is 3 months?
pleonasm
Posts: 30
Joined: Tue Oct 15, 2019 11:59 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by pleonasm »

Semantics wrote: Tue Aug 04, 2020 1:34 am I think you missed a couple of zeros. I get 237800% for SQQQ (~2500x).
Thereum wrote: Tue Aug 04, 2020 1:37 am pleonasm,

I understand why you are getting your result. You need to allow monthly rebalancing for the short-only portfolios.

Make a portfolio with -100% SQQQ and 100% CASHX and rebalance monthly. You should see the right results with this.

Oh wow. I see what you guys are doing.

You are realizing all profits monthly and pushing 100% back into a new short in the backtest. This is giving 120 months of 100% sequence of returns data with a snowballing pile of cash for every short. This however is not shrinking exposure at any point if you do it correctly. If you are doing this with options you need to open net new positions monthly (appears you are), but this retards the gains of the older positions as the arithmetic value gets smaller. To get anything close to your PV values you need to close all positions and reopen all positions with as close to -100 delta as possible via short calls or long puts.

The arithmetic will still show that the relation between SQQQ and TQQQ that the larger share price out performs inside a period in relation to the other.

You are not going to realize anything close to these backtests via options. To do this properly (and short directly and monthly) you would have to short in a taxable account which would give you monstrous tax drag.
Semantics
Posts: 185
Joined: Tue Mar 10, 2020 1:42 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Semantics »

pleonasm wrote: Tue Aug 04, 2020 9:21 am
Semantics wrote: Tue Aug 04, 2020 1:34 am I think you missed a couple of zeros. I get 237800% for SQQQ (~2500x).
Thereum wrote: Tue Aug 04, 2020 1:37 am pleonasm,

I understand why you are getting your result. You need to allow monthly rebalancing for the short-only portfolios.

Make a portfolio with -100% SQQQ and 100% CASHX and rebalance monthly. You should see the right results with this.

Oh wow. I see what you guys are doing.

You are realizing all profits monthly and pushing 100% back into a new short in the backtest. This is giving 120 months of 100% sequence of returns data with a snowballing pile of cash for every short. This however is not shrinking exposure at any point if you do it correctly. If you are doing this with options you need to open net new positions monthly (appears you are), but this retards the gains of the older positions as the arithmetic value gets smaller. To get anything close to your PV values you need to close all positions and reopen all positions with as close to -100 delta as possible via short calls or long puts.
Yup, and even if you hold to expiry so delta is irrelevant, the strike price of the short call is likely ITM when it's opened, so if SQQQ doesn't drop enough to cover that gap you'll owe money on the spread when the options expire/are closed. That is guaranteed to make this strategy do worse than straight-up shorting (assuming infinite margin). It will probably do better than TQQQ, but as always, that's because it takes on more risk - losses on the call spread will almost always be higher than what TQQQ would lose, and backtests only have ~120 months in a bull market. A couple dozen months with significant losses is not enough of a sample size to properly capture the worst cases, which are when a big drop coincides with the rebalancing date.
Thereum
Posts: 62
Joined: Sun Jun 14, 2020 9:05 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Thereum »

pleonasm wrote: Tue Aug 04, 2020 9:21 am
Semantics wrote: Tue Aug 04, 2020 1:34 am I think you missed a couple of zeros. I get 237800% for SQQQ (~2500x).
Thereum wrote: Tue Aug 04, 2020 1:37 am pleonasm,

I understand why you are getting your result. You need to allow monthly rebalancing for the short-only portfolios.

Make a portfolio with -100% SQQQ and 100% CASHX and rebalance monthly. You should see the right results with this.

Oh wow. I see what you guys are doing.

You are realizing all profits monthly and pushing 100% back into a new short in the backtest. This is giving 120 months of 100% sequence of returns data with a snowballing pile of cash for every short. This however is not shrinking exposure at any point if you do it correctly. If you are doing this with options you need to open net new positions monthly (appears you are), but this retards the gains of the older positions as the arithmetic value gets smaller. To get anything close to your PV values you need to close all positions and reopen all positions with as close to -100 delta as possible via short calls or long puts.

The arithmetic will still show that the relation between SQQQ and TQQQ that the larger share price out performs inside a period in relation to the other.

You are not going to realize anything close to these backtests via options. To do this properly (and short directly and monthly) you would have to short in a taxable account which would give you monstrous tax drag.
Well, I have backtested options strategies using ORATS and found exceptional returns. You would probably lose 5% due to commissions and slippage. Still exceptional.

And again, going long leveraged ETFs is fundamentally flawed, even though it has worked and will likely continue to work. It's still a flawed strategy due to the volatility decay of leveraged ETFs. Shorting leveraged ETFs is betting against a flawed strategy. There is alpha built into the approach.
pleonasm
Posts: 30
Joined: Tue Oct 15, 2019 11:59 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by pleonasm »

Thereum wrote: Tue Aug 04, 2020 11:51 am
pleonasm wrote: Tue Aug 04, 2020 9:21 am
Semantics wrote: Tue Aug 04, 2020 1:34 am I think you missed a couple of zeros. I get 237800% for SQQQ (~2500x).
Thereum wrote: Tue Aug 04, 2020 1:37 am pleonasm,

I understand why you are getting your result. You need to allow monthly rebalancing for the short-only portfolios.

Make a portfolio with -100% SQQQ and 100% CASHX and rebalance monthly. You should see the right results with this.

Oh wow. I see what you guys are doing.

You are realizing all profits monthly and pushing 100% back into a new short in the backtest. This is giving 120 months of 100% sequence of returns data with a snowballing pile of cash for every short. This however is not shrinking exposure at any point if you do it correctly. If you are doing this with options you need to open net new positions monthly (appears you are), but this retards the gains of the older positions as the arithmetic value gets smaller. To get anything close to your PV values you need to close all positions and reopen all positions with as close to -100 delta as possible via short calls or long puts.
Well, I have backtested options strategies using ORATS and found exceptional returns. You would probably lose 5% due to commissions and slippage. Still exceptional.

And again, going long leveraged ETFs is fundamentally flawed, even though it has worked and will likely continue to work. It's still a flawed strategy due to the volatility decay of leveraged ETFs. Shorting leveraged ETFs is betting against a flawed strategy. There is alpha built into the approach.
ORATS with Binomial pricing and holding to expiry?

I'm showing a Buying Power Effect of -$5490 for your Jan SQQQ positions.

Your Profit @ $5 should be in the ballpark of $800, @$4 $1200.

From 5.67, you need 12 straight +1% days of QQQ to get to $4. In that same period TQQQ would gain 124 > 176, or 42%. SQQQ would return 33%, and your position will return 22%. This is if everything goes 100% correct for you. If we get only 5 net +1% QQQ days, SQQQ @ 5 & TQQQ at 139.56, SQQQ returns 9%, TQQQ returns 12.5%, you return $800 or 14.6%

Of course, it will not happen this way. But because these are rebalanced daily, positive sequence of returns accelerates the TQQQ because of the arithmetic value faster than SQQQ. Negative sequence of returns runs over your long put and destroys your value proposition. SQQQ @ 7.40 (+30.5%) (10 net -1% QQQ days) puts TQQQ at 94.27 (-24%) and your loss at -1900 or -34%.

Your Jan position is giving you approx -500 deltas because of the spread (you have a 1100 share buying power reduction). You will need to bleed your short calls all the way dry to get close to your profits I am citing, but at that point you will experience the 1200 share leverage as the intrinsics go to zero.

You are losing more than 5% slippage to the greeks and spreads and commissions. You have lower than 1x leverage in the beginning of the trade because of the nature of the spread, and then higher than 1x at the end. Your December position would create 2x leverage if ITM in the last week before expiring, giving you a huge risk of sequence of returns.

You would need to do the option analysis monthly and deploy the new capital to get this to work for the return you are seeking.
hilink73
Posts: 537
Joined: Tue Sep 20, 2016 3:29 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by hilink73 »

Thereum wrote: Tue Aug 04, 2020 11:51 am It's still a flawed strategy due to the volatility decay of leveraged ETFs.
We have discussed volatility decay on numerous pages here and finally concluded that it isn't so much of an issue, really.
At least, it goes both ways.
Impatience
Posts: 275
Joined: Thu Jul 23, 2020 3:15 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Impatience »

hilink73 wrote: Tue Aug 04, 2020 2:31 pm
Thereum wrote: Tue Aug 04, 2020 11:51 am It's still a flawed strategy due to the volatility decay of leveraged ETFs.
We have discussed volatility decay on numerous pages here and finally concluded that it isn't so much of an issue, really.
At least, it goes both ways.
Right. ALL assets experience “volatility decay” whether it’s a vanguard mutual fund share or UPRO. It’s a fact of math not of a particular financial product.
User avatar
mrspock
Posts: 1395
Joined: Tue Feb 13, 2018 2:49 am
Location: Vulcan

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by mrspock »

Meaty wrote: Mon Aug 03, 2020 11:12 am
HawkeyePierce wrote: Mon Aug 03, 2020 12:53 am I'm just about at the one year mark of diving into the EDV flavor of the Adventure. In that time, my allocation is up 49.57%.

(43/57 UPRO/EDV, quarterly rebals, entirely in Roth space)
Started a year ago this month. 55/45 UPRO TMF. I’m up 46%.
I'm doing the 50/50 version of this w/ quarterly rebalance.

I'm up similar, I've honestly lost count....but my original investment is up north of 50% now (started Feb 2019). I've been so impressed, I tripled my investment in this, to 5% of my portfolio and it's been pretty fun to watch. I'm doing this in my ROTH 401k, so I'll consider it a great success if I get audited one day by the IRS as they are puzzled how I got $1-2m in there :D .

I consider this strategy a perfect play for a ROTH account, as you are comp'd for the risk to a degree since the money you make in that account is worth more than anywhere else due to the tax treatment. Kind of juices the gains a bit.
guyinlaw
Posts: 727
Joined: Wed Jul 03, 2019 9:54 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by guyinlaw »

If anyone is investing >$10k and >5% of your portfolio consider PSLDX in your IRA account. It is approximately ~2X leveraged stocks/bonds.

We are heading to the end of 40year bond bull market. As TMF is hitting it's roof, this stratergy is getting riskier.
Time is your friend; impulse is your enemy. - John C. Bogle
potatopancake
Posts: 38
Joined: Fri Jul 13, 2018 4:38 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by potatopancake »

What is TMF's roof?
SVT
Posts: 368
Joined: Mon Oct 13, 2008 8:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by SVT »

guyinlaw wrote: Tue Aug 04, 2020 3:33 pm If anyone is investing >$10k and >5% of your portfolio consider PSLDX in your IRA account. It is approximately ~2X leveraged stocks/bonds.
UPRO/TMF is getting close to 20% of my total net worth now (Roth IRA). I also currently have about 35% of my net worth in PSLDX (Traditional IRA).
SVT
Posts: 368
Joined: Mon Oct 13, 2008 8:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by SVT »

mrspock wrote: Tue Aug 04, 2020 3:24 pm
Meaty wrote: Mon Aug 03, 2020 11:12 am
HawkeyePierce wrote: Mon Aug 03, 2020 12:53 am I'm just about at the one year mark of diving into the EDV flavor of the Adventure. In that time, my allocation is up 49.57%.

(43/57 UPRO/EDV, quarterly rebals, entirely in Roth space)
Started a year ago this month. 55/45 UPRO TMF. I’m up 46%.
I'm doing the 50/50 version of this w/ quarterly rebalance.

I'm up similar, I've honestly lost count....but my original investment is up north of 50% now (started Feb 2019).
Likely way north of 50%. Probably well over 100%.
ChrisBenn
Posts: 455
Joined: Mon Aug 05, 2019 7:56 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by ChrisBenn »

Thereum wrote: Tue Aug 04, 2020 11:51 am
pleonasm wrote: Tue Aug 04, 2020 9:21 am
Semantics wrote: Tue Aug 04, 2020 1:34 am I think you missed a couple of zeros. I get 237800% for SQQQ (~2500x).
Thereum wrote: Tue Aug 04, 2020 1:37 am pleonasm,

I understand why you are getting your result. You need to allow monthly rebalancing for the short-only portfolios.

Make a portfolio with -100% SQQQ and 100% CASHX and rebalance monthly. You should see the right results with this.

Oh wow. I see what you guys are doing.

You are realizing all profits monthly and pushing 100% back into a new short in the backtest. This is giving 120 months of 100% sequence of returns data with a snowballing pile of cash for every short. This however is not shrinking exposure at any point if you do it correctly. If you are doing this with options you need to open net new positions monthly (appears you are), but this retards the gains of the older positions as the arithmetic value gets smaller. To get anything close to your PV values you need to close all positions and reopen all positions with as close to -100 delta as possible via short calls or long puts.

The arithmetic will still show that the relation between SQQQ and TQQQ that the larger share price out performs inside a period in relation to the other.

You are not going to realize anything close to these backtests via options. To do this properly (and short directly and monthly) you would have to short in a taxable account which would give you monstrous tax drag.
Well, I have backtested options strategies using ORATS and found exceptional returns. You would probably lose 5% due to commissions and slippage. Still exceptional.

And again, going long leveraged ETFs is fundamentally flawed, even though it has worked and will likely continue to work. It's still a flawed strategy due to the volatility decay of leveraged ETFs. Shorting leveraged ETFs is betting against a flawed strategy. There is alpha built into the approach.
If your target is the "alpha" from shorting leveraged ETF's you could short both pairs (TQQQ/SQQQ). The 3x equity exposure would cancel so you would only be collecting on said inefficiencies. This would be a great test to see if the presumed structural inefficiency returns are more than borrowing fees, margin fees, and taxes -- at much less risk than shorting just sqqq, and without paying the optionality premium.
Semantics
Posts: 185
Joined: Tue Mar 10, 2020 1:42 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Semantics »

guyinlaw wrote: Tue Aug 04, 2020 3:33 pm If anyone is investing >$10k and >5% of your portfolio consider PSLDX in your IRA account. It is approximately ~2X leveraged stocks/bonds.

We are heading to the end of 40year bond bull market. As TMF is hitting it's roof, this stratergy is getting riskier.
In a taxable account one might consider BTAL. From 2012-18 (flat 30 year yield) a UPRO/BTAL portfolio has better Sharpe ratio than UPRO/TMF. BTAL won't generate any returns, but it has a strong negative correlation to stocks, so it helps more to reduce volatility. Needs to be levered up manually to get the full effect though.
Semantics
Posts: 185
Joined: Tue Mar 10, 2020 1:42 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Semantics »

Thereum wrote: Tue Aug 04, 2020 11:51 am
pleonasm wrote: Tue Aug 04, 2020 9:21 am
Semantics wrote: Tue Aug 04, 2020 1:34 am I think you missed a couple of zeros. I get 237800% for SQQQ (~2500x).
Thereum wrote: Tue Aug 04, 2020 1:37 am pleonasm,

I understand why you are getting your result. You need to allow monthly rebalancing for the short-only portfolios.

Make a portfolio with -100% SQQQ and 100% CASHX and rebalance monthly. You should see the right results with this.

Oh wow. I see what you guys are doing.

You are realizing all profits monthly and pushing 100% back into a new short in the backtest. This is giving 120 months of 100% sequence of returns data with a snowballing pile of cash for every short. This however is not shrinking exposure at any point if you do it correctly. If you are doing this with options you need to open net new positions monthly (appears you are), but this retards the gains of the older positions as the arithmetic value gets smaller. To get anything close to your PV values you need to close all positions and reopen all positions with as close to -100 delta as possible via short calls or long puts.

The arithmetic will still show that the relation between SQQQ and TQQQ that the larger share price out performs inside a period in relation to the other.

You are not going to realize anything close to these backtests via options. To do this properly (and short directly and monthly) you would have to short in a taxable account which would give you monstrous tax drag.
Well, I have backtested options strategies using ORATS and found exceptional returns. You would probably lose 5% due to commissions and slippage. Still exceptional.

And again, going long leveraged ETFs is fundamentally flawed, even though it has worked and will likely continue to work. It's still a flawed strategy due to the volatility decay of leveraged ETFs. Shorting leveraged ETFs is betting against a flawed strategy. There is alpha built into the approach.
There's no alpha built into the approach. It takes on more risk to produce the excess returns.
jarjarM
Posts: 431
Joined: Mon Jul 16, 2018 1:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by jarjarM »

SVT wrote: Tue Aug 04, 2020 4:34 pm
mrspock wrote: Tue Aug 04, 2020 3:24 pm
Meaty wrote: Mon Aug 03, 2020 11:12 am
HawkeyePierce wrote: Mon Aug 03, 2020 12:53 am I'm just about at the one year mark of diving into the EDV flavor of the Adventure. In that time, my allocation is up 49.57%.

(43/57 UPRO/EDV, quarterly rebals, entirely in Roth space)
Started a year ago this month. 55/45 UPRO TMF. I’m up 46%.
I'm doing the 50/50 version of this w/ quarterly rebalance.

I'm up similar, I've honestly lost count....but my original investment is up north of 50% now (started Feb 2019).
Likely way north of 50%. Probably well over 100%.
Very unlikely to be over 100% if there's no additional $$$ added in. Depending on the exact date one starts this, UPRO could be still underwater (after 7/2019 entry date) or just ~20% above 2/2019 price level. While TMF is ~100% return since the original adventure started, UPRO return is subpar due to the march drawdown. Volatility decay is very real with the 3x LETF.

Personally I'm up ~70%, started around March 2019.
Thereum
Posts: 62
Joined: Sun Jun 14, 2020 9:05 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Thereum »

ChrisBenn wrote: Tue Aug 04, 2020 5:14 pm If your target is the "alpha" from shorting leveraged ETF's you could short both pairs (TQQQ/SQQQ). The 3x equity exposure would cancel so you would only be collecting on said inefficiencies. This would be a great test to see if the presumed structural inefficiency returns are more than borrowing fees, margin fees, and taxes -- at much less risk than shorting just sqqq, and without paying the optionality premium.
I have backtested this approach and you do get alpha. Every single year of -50/-50 has been a winner except 2019, which was a strongly trending market. I believe more frequent rebalancing would have fixed this. If you do this strategy on something like JNUG and DUST, the results are poor, but still profitable. Again, I believe that frequent rebalancing would help, but I haven't studied this approach closely.

Another approach is to just short Hedgefundie's portfolio and short the complementary inverse ETFs. You get amazing results with this approach. I don't fully trust PortfolioVisualizer, but its calculations are saying I could lever this strategy 50:1 without blowing up. I am sure quantitative firms are already doing this, with proper risk management and near instantaneous hedging. As a retail investor, I will stick with options so that I can define my risk.
jarjarM
Posts: 431
Joined: Mon Jul 16, 2018 1:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by jarjarM »

guyinlaw wrote: Tue Aug 04, 2020 3:33 pm If anyone is investing >$10k and >5% of your portfolio consider PSLDX in your IRA account. It is approximately ~2X leveraged stocks/bonds.

We are heading to the end of 40year bond bull market. As TMF is hitting it's roof, this stratergy is getting riskier.
The original argument from hedgefundie is that TMF is strictly there to be the counter balance during significant market disruption which impacts UPRO greatly as we had seen during the march time frame. Obviously when to put to practice, TMF is the main driving force for the return of this portfolio since Feb 2019. Some would argue TMF still can be profitable given potential negative interest scenario, one would have to decide if they're willing to bet on that with a portion of this strategy. As others have pointed out, there are other potential pairing that could serve as counter balance during significant UPRO drawdown.

P.S. I too have a significant exposure to PSLDX, which is why I'm watching the interest rate trend with great interest. :beer
Thereum
Posts: 62
Joined: Sun Jun 14, 2020 9:05 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Thereum »

jarjarM wrote: Tue Aug 04, 2020 5:29 pm Very unlikely to be over 100% if there's no additional $$$ added in. Depending on the exact date one starts this, UPRO could be still underwater (after 7/2019 entry date) or just ~20% above 2/2019 price level. While TMF is ~100% return since the original adventure started, UPRO return is subpar due to the march drawdown. Volatility decay is very real with the 3x LETF.

Personally I'm up ~70%, started around March 2019.
I am glad you are acknowledging the volatility decay in going long on leveraged ETFs.

Also, what happens when rates rise or investors decide they'd rather hold cash during a crash than LTTs yielding less than 1%? We began to see an exit from LTTs during the March crash until the Fed intervened. Some people doing Hedgefundie's approach on margin blew up when this happened.

As I said before, Hedgefundie's approach is pure recency bias and over-fitting. There is no guarantee that TMF will continue to serve as a free put option during crashes.
User avatar
mrspock
Posts: 1395
Joined: Tue Feb 13, 2018 2:49 am
Location: Vulcan

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by mrspock »

SVT wrote: Tue Aug 04, 2020 4:34 pm
mrspock wrote: Tue Aug 04, 2020 3:24 pm
Meaty wrote: Mon Aug 03, 2020 11:12 am
HawkeyePierce wrote: Mon Aug 03, 2020 12:53 am I'm just about at the one year mark of diving into the EDV flavor of the Adventure. In that time, my allocation is up 49.57%.

(43/57 UPRO/EDV, quarterly rebals, entirely in Roth space)
Started a year ago this month. 55/45 UPRO TMF. I’m up 46%.
I'm doing the 50/50 version of this w/ quarterly rebalance.

I'm up similar, I've honestly lost count....but my original investment is up north of 50% now (started Feb 2019).
Likely way north of 50%. Probably well over 100%.
Yes sorry, over 100% :D . My bad hehe, I added money once to that ROTH account this year, but I’m not counting that. On my original investment it was 24k and now I’m well above 50k. I’ve also been super lucky in the timing of the rebalancing... just how the dates worked out (I do rebalancing exactly on first trading day of new quarter).
Last edited by mrspock on Tue Aug 04, 2020 5:49 pm, edited 1 time in total.
jarjarM
Posts: 431
Joined: Mon Jul 16, 2018 1:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by jarjarM »

Thereum wrote: Tue Aug 04, 2020 5:40 pm
jarjarM wrote: Tue Aug 04, 2020 5:29 pm Very unlikely to be over 100% if there's no additional $$$ added in. Depending on the exact date one starts this, UPRO could be still underwater (after 7/2019 entry date) or just ~20% above 2/2019 price level. While TMF is ~100% return since the original adventure started, UPRO return is subpar due to the march drawdown. Volatility decay is very real with the 3x LETF.

Personally I'm up ~70%, started around March 2019.
I am glad you are acknowledging the volatility decay in going long on leveraged ETFs.

Also, what happens when rates rise or investors decide they'd rather hold cash during a crash than LTTs yielding less than 1%? We began to see an exit from LTTs during the March crash until the Fed intervened. Some people doing Hedgefundie's approach on margin blew up when this happened.

As I said before, Hedgefundie's approach is pure recency bias and over-fitting. There is no guarantee that TMF will continue to serve as a free put option during crashes.
The decay was discussed in the original and this thread so I think those of us who actually followed since the beginning is well aware of the risk (hence no one is holding UPRO with margin I hope). At this point, LTT is still the way most institutional players will go when the crash happens, at least until USD loses reserve currency status, which is whole other discussion. I do remember a poster's margin blew up in the depth of the march drawdown when LTT and SP500 got hit. There's no free lunch and every strategy will have some risk.

Also, as Semantic pointed out in a couple of places, the problem with your strategy is tax. Since shorting is generally not allowed in IRA, one would need to do this in taxable account and pay ~50% in tax if in top fed/CA tax bracket. That's a BIG tax drag.
jarjarM
Posts: 431
Joined: Mon Jul 16, 2018 1:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by jarjarM »

mrspock wrote: Tue Aug 04, 2020 5:45 pm
SVT wrote: Tue Aug 04, 2020 4:34 pm
mrspock wrote: Tue Aug 04, 2020 3:24 pm
Meaty wrote: Mon Aug 03, 2020 11:12 am
HawkeyePierce wrote: Mon Aug 03, 2020 12:53 am I'm just about at the one year mark of diving into the EDV flavor of the Adventure. In that time, my allocation is up 49.57%.

(43/57 UPRO/EDV, quarterly rebals, entirely in Roth space)
Started a year ago this month. 55/45 UPRO TMF. I’m up 46%.
I'm doing the 50/50 version of this w/ quarterly rebalance.

I'm up similar, I've honestly lost count....but my original investment is up north of 50% now (started Feb 2019).
Likely way north of 50%. Probably well over 100%.
Yes sorry, over 100% :D . My bad hehe, I added money once to that ROTH account this year, but I’m not counting that. On my original investment it was 24k and now I’m well above 50k. I’ve also been super lucky in the timing of the rebalancing... just how the dates worked out (I do rebalancing exactly on first trading day of new quarter).
Congrats, my rebalancing were not as lucky.
RocketShipTech
Posts: 679
Joined: Sat Jun 13, 2020 10:08 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by RocketShipTech »

Thereum wrote: Tue Aug 04, 2020 5:40 pm
jarjarM wrote: Tue Aug 04, 2020 5:29 pm Very unlikely to be over 100% if there's no additional $$$ added in. Depending on the exact date one starts this, UPRO could be still underwater (after 7/2019 entry date) or just ~20% above 2/2019 price level. While TMF is ~100% return since the original adventure started, UPRO return is subpar due to the march drawdown. Volatility decay is very real with the 3x LETF.

Personally I'm up ~70%, started around March 2019.
I am glad you are acknowledging the volatility decay in going long on leveraged ETFs.

Also, what happens when rates rise or investors decide they'd rather hold cash during a crash than LTTs yielding less than 1%? We began to see an exit from LTTs during the March crash until the Fed intervened. Some people doing Hedgefundie's approach on margin blew up when this happened.

As I said before, Hedgefundie's approach is pure recency bias and over-fitting. There is no guarantee that TMF will continue to serve as a free put option during crashes.
And so what do you think the Fed will do the next time investors begin to "exit from LTTs"?
SVT
Posts: 368
Joined: Mon Oct 13, 2008 8:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by SVT »

jarjarM wrote: Tue Aug 04, 2020 5:29 pm
SVT wrote: Tue Aug 04, 2020 4:34 pm
mrspock wrote: Tue Aug 04, 2020 3:24 pm
Meaty wrote: Mon Aug 03, 2020 11:12 am
HawkeyePierce wrote: Mon Aug 03, 2020 12:53 am I'm just about at the one year mark of diving into the EDV flavor of the Adventure. In that time, my allocation is up 49.57%.

(43/57 UPRO/EDV, quarterly rebals, entirely in Roth space)
Started a year ago this month. 55/45 UPRO TMF. I’m up 46%.
I'm doing the 50/50 version of this w/ quarterly rebalance.

I'm up similar, I've honestly lost count....but my original investment is up north of 50% now (started Feb 2019).
Likely way north of 50%. Probably well over 100%.
Very unlikely to be over 100% if there's no additional $$$ added in. Depending on the exact date one starts this, UPRO could be still underwater (after 7/2019 entry date) or just ~20% above 2/2019 price level. While TMF is ~100% return since the original adventure started, UPRO return is subpar due to the march drawdown. Volatility decay is very real with the 3x LETF.

Personally I'm up ~70%, started around March 2019.
I started Feb 2019, just like spock, I'm up 135%. Haven't added a penny. Exact asset allocation and timing of rebalancing matters and I believe I did get a little lucky on keeping 40/60 longer than OP and others and also didn't rebalance exactly quarterly which I think worked in my favor. Maybe I got more lucky than I thought and spock is in fact up "only" 70%.

ETA: after today, up 140% not 135%
Last edited by SVT on Tue Aug 04, 2020 6:05 pm, edited 1 time in total.
jarjarM
Posts: 431
Joined: Mon Jul 16, 2018 1:21 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by jarjarM »

SVT wrote: Tue Aug 04, 2020 5:58 pm
jarjarM wrote: Tue Aug 04, 2020 5:29 pm
SVT wrote: Tue Aug 04, 2020 4:34 pm
mrspock wrote: Tue Aug 04, 2020 3:24 pm
Meaty wrote: Mon Aug 03, 2020 11:12 am

Started a year ago this month. 55/45 UPRO TMF. I’m up 46%.
I'm doing the 50/50 version of this w/ quarterly rebalance.

I'm up similar, I've honestly lost count....but my original investment is up north of 50% now (started Feb 2019).
Likely way north of 50%. Probably well over 100%.
Very unlikely to be over 100% if there's no additional $$$ added in. Depending on the exact date one starts this, UPRO could be still underwater (after 7/2019 entry date) or just ~20% above 2/2019 price level. While TMF is ~100% return since the original adventure started, UPRO return is subpar due to the march drawdown. Volatility decay is very real with the 3x LETF.

Personally I'm up ~70%, started around March 2019.
I started Feb 2019, just like spock, I'm up 135%. Haven't added a penny. Exact asset allocation and timing of rebalancing matters and I believe I did get a little lucky on keeping 40/60 longer than OP and others and also didn't rebalance exactly quarterly which I think worked in my favor. Maybe I got more lucky than I thought and spock is in fact up "only" 70%.
Ah I see, congrats. I do think because the leverage and the recent volatility, the exact rebalance date/price makes quite a bit of difference.
SVT
Posts: 368
Joined: Mon Oct 13, 2008 8:56 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by SVT »

mrspock wrote: Tue Aug 04, 2020 5:45 pm
SVT wrote: Tue Aug 04, 2020 4:34 pm
mrspock wrote: Tue Aug 04, 2020 3:24 pm
Meaty wrote: Mon Aug 03, 2020 11:12 am
HawkeyePierce wrote: Mon Aug 03, 2020 12:53 am I'm just about at the one year mark of diving into the EDV flavor of the Adventure. In that time, my allocation is up 49.57%.

(43/57 UPRO/EDV, quarterly rebals, entirely in Roth space)
Started a year ago this month. 55/45 UPRO TMF. I’m up 46%.
I'm doing the 50/50 version of this w/ quarterly rebalance.

I'm up similar, I've honestly lost count....but my original investment is up north of 50% now (started Feb 2019).
Likely way north of 50%. Probably well over 100%.
Yes sorry, over 100% :D . My bad hehe, I added money once to that ROTH account this year, but I’m not counting that. On my original investment it was 24k and now I’m well above 50k. I’ve also been super lucky in the timing of the rebalancing... just how the dates worked out (I do rebalancing exactly on first trading day of new quarter).
Yep, cool, that's what I thought.
Semantics
Posts: 185
Joined: Tue Mar 10, 2020 1:42 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Semantics »

Thereum wrote: Tue Aug 04, 2020 5:36 pm
ChrisBenn wrote: Tue Aug 04, 2020 5:14 pm If your target is the "alpha" from shorting leveraged ETF's you could short both pairs (TQQQ/SQQQ). The 3x equity exposure would cancel so you would only be collecting on said inefficiencies. This would be a great test to see if the presumed structural inefficiency returns are more than borrowing fees, margin fees, and taxes -- at much less risk than shorting just sqqq, and without paying the optionality premium.
I have backtested this approach and you do get alpha. Every single year of -50/-50 has been a winner except 2019, which was a strongly trending market. I believe more frequent rebalancing would have fixed this. If you do this strategy on something like JNUG and DUST, the results are poor, but still profitable. Again, I believe that frequent rebalancing would help, but I haven't studied this approach closely.
If you rebalance daily it's identical to long ETFs, since you maintain identical exposure. With monthly rebalancing, I still don't see why it's so obvious that there is additional alpha - you're exposed to more risk, how can you assume there's alpha without quantifying that risk? The additional risk doesn't show up in Portfolio Visualizer because 120 months of data is too small a sample size to capture the worst outcomes. PV will happily ignore the case where your portfolio goes negative mid-month as long as things recover, because that part show up in the data.

I've pointed out before that the inverse of volatility decay is just swing trading. If that's a good source of alpha, why doesn't everyone just swing trade and get rich? Then there's no need to mess around with margin or shorting.
BogleBobby
Posts: 38
Joined: Tue Jul 30, 2019 9:56 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by BogleBobby »

Regarding the question "how much more return does TMF have with interest rates so low?", I'll point out a few things:

This was a concern when this thread first started and TMF has returned 64% over the last year.

Interest rates have been falling for a long time (hundreds of years if you go back far enough) and part of the conviction of this strategy is that they will continue to fall or stay low. If you think they will start going up significantly, this strategy isn't for you.

See an article here about the fall of interest rates over time: https://www.visualcapitalist.com/700-ye ... est-rates/

TMF still has plenty of return left because interest rates can still fall further. We still have positive interest rates on the 20-year and 30-year. Interest rates can go negative.

However, you don't necessarily need a big return from TMF for this strategy to succeed. Instead, you need it to continue to be negatively correlated with equities during times of crisis.

Interest rates being really low is a good thing for this strategy because long term bond funds are more sensitive to interest rate changes when interest rates are low due to bond convexity. You want TMF to be more volatile to complement your equity holdings. See the article here (which has been linked before in this discussion): https://portfoliocharts.com/2019/05/27/ ... convexity/
Thereum
Posts: 62
Joined: Sun Jun 14, 2020 9:05 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Thereum »

BogleBobby,

You make some very good points. I am beginning to reconsider my stance.

I guess my main concern was choppiness in rates, which could hurt TMF. This is why I was thinking of going short TMV, which is a strategy that would benefit from choppiness. But your argument makes me think that holding TMF is still justified.
tomphilly
Posts: 73
Joined: Wed Aug 05, 2020 11:29 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by tomphilly »

Hi all, I'm new here and I have been following this thread and other discussions like it elsewhere with great interest for a while. I've been employing a UPRO/TMF 60/40 ratio since mid 2019 with good success. I share the concern of some here about interest rates and their effect on TMF in the long term. But I also acknowledge that there's a strong chance they won't rise any time soon and historically interest rates have been dropping for centuries - however, to be practical, this strategy needs to be effective now and during this decade, not during the next decade or when we look back over the past century of investment strategies on boggleheads.org in the year 2100. Clearly no one would hang on to this strategy over certain decades in the last century (60's to 80's) and the 2020's could become one of those decades.

When I think about it I have more certainty that the SP500 will rise than LTT's yields will drop over the next decade. So the result of this thought experiment is to ask what can be done to feel more comfortable with the volatility hedge. A possible rise in interest rates is just a known unknown that will affect TMF. My concern is there are unknown unknowns that could cause LTT's to deteriorate:

1) During the March crash the volatility hedge started to break down on several trading days, leading to some eye-watering drawdowns for me - I don't fully understand why, but I believe the aggressive fed intervention played some part in this strategy not being completely wiped out.
2) Today, TMF is down 3% in response to a Fed debt sale - is this a once off, or will it continue?
3) What happens to LTT's if the US dollar is no longer the world's reserve currency?
4) What happens to LTT's if we returned to the gold standard? (Not a joke - it is a policy position of the new fed candidate)

I know you might think, "Well, we can dump or adjust this strategy before these things take place", but that seems a lot like throwing a frog in boiling water (frog jumps out & survives) versus putting a frog in cold water and heating it up (frog slowly dies).

Note that I'm not nearly as financially or macroeconomically savvy as many of you here, I'm more "financially curious" as someone else put it. With my limited savviness, I'm looking at alternative volatility hedges like BTAL, TYD and UST, to see if the strategy could still be effective with a basket like 15% TMF, 15% TYD & 10% BTAL.
RocketShipTech
Posts: 679
Joined: Sat Jun 13, 2020 10:08 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by RocketShipTech »

tomphilly wrote: Wed Aug 05, 2020 12:15 pm
I know you might think, "Well, we can dump or adjust this strategy before these things take place", but that seems a lot like throwing a frog in boiling water (frog jumps out & survives) versus putting a frog in cold water and heating it up (frog slowly dies).
This is a myth.

https://io9.gizmodo.com/frogs-are-not-o ... 1493614589
User avatar
Meaty
Posts: 814
Joined: Mon Jul 22, 2013 7:35 pm
Location: Florida

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Meaty »

Given the Fed has indicated they want to increase inflation - what does that mean for this strategy? Rates stay low for longer but not sure the implications for this risk parity play
"Discipline equals Freedom" - Jocko Willink
User avatar
cos
Posts: 288
Joined: Fri Aug 23, 2019 7:34 pm
Location: Boston
Contact:

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by cos »

Meaty wrote: Wed Aug 05, 2020 4:43 pm Given the Fed has indicated they want to increase inflation - what does that mean for this strategy? Rates stay low for longer but not sure the implications for this risk parity play
When and where did they indicate this? Aren't they still targeting 2%? As I understand it, they're still desperately battling deflation.
User avatar
Meaty
Posts: 814
Joined: Mon Jul 22, 2013 7:35 pm
Location: Florida

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Meaty »

cos wrote: Wed Aug 05, 2020 5:22 pm
Meaty wrote: Wed Aug 05, 2020 4:43 pm Given the Fed has indicated they want to increase inflation - what does that mean for this strategy? Rates stay low for longer but not sure the implications for this risk parity play
When and where did they indicate this? Aren't they still targeting 2%? As I understand it, they're still desperately battling deflation.
Yesterday

https://www.google.com/amp/s/www.cnbc.c ... -soon.html
"Discipline equals Freedom" - Jocko Willink
User avatar
cos
Posts: 288
Joined: Fri Aug 23, 2019 7:34 pm
Location: Boston
Contact:

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by cos »

Meaty wrote: Wed Aug 05, 2020 5:39 pm
cos wrote: Wed Aug 05, 2020 5:22 pm
Meaty wrote: Wed Aug 05, 2020 4:43 pm Given the Fed has indicated they want to increase inflation - what does that mean for this strategy? Rates stay low for longer but not sure the implications for this risk parity play
When and where did they indicate this? Aren't they still targeting 2%? As I understand it, they're still desperately battling deflation.
Yesterday

https://www.google.com/amp/s/www.cnbc.c ... -soon.html
Oh, wow, thanks for the link! So it looks like they won't do anything until the 2% target is reached. The author surmises, "With inflation now closer to 1% and the jobless rate higher than it’s been since the Great Depression, the likelihood is that the Fed could need years to hit its targets," and I'm inclined to agree. This shouldn't be a problem in the near future, but it could be a problem a bit further down the line. Fortunately, the Fed indicated that their target remains 2%, they just plan to be more lax about reigning it in whenever it surpasses 2%.
User avatar
Meaty
Posts: 814
Joined: Mon Jul 22, 2013 7:35 pm
Location: Florida

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Meaty »

cos wrote: Wed Aug 05, 2020 5:58 pm
Meaty wrote: Wed Aug 05, 2020 5:39 pm
cos wrote: Wed Aug 05, 2020 5:22 pm
Meaty wrote: Wed Aug 05, 2020 4:43 pm Given the Fed has indicated they want to increase inflation - what does that mean for this strategy? Rates stay low for longer but not sure the implications for this risk parity play
When and where did they indicate this? Aren't they still targeting 2%? As I understand it, they're still desperately battling deflation.
Yesterday

https://www.google.com/amp/s/www.cnbc.c ... -soon.html
Oh, wow, thanks for the link! So it looks like they won't do anything until the 2% target is reached. The author surmises, "With inflation now closer to 1% and the jobless rate higher than it’s been since the Great Depression, the likelihood is that the Fed could need years to hit its targets," and I'm inclined to agree. This shouldn't be a problem in the near future, but it could be a problem a bit further down the line. Fortunately, the Fed indicated that their target remains 2%, they just plan to be more lax about reigning it in whenever it surpasses 2%.
Thanks for the quick look. So do we know what the impact of higher inflation would be to this strategy
"Discipline equals Freedom" - Jocko Willink
DeepLearner
Posts: 1
Joined: Wed Aug 05, 2020 8:01 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by DeepLearner »

IntolerableOptimism wrote: Mon Aug 03, 2020 11:16 am
This strategy isn't a free lunch and requires a lot of babysitting. Essentially, a full-time leveraged, risk parity strategy will underperform a non-leveraged 100% stock portfolio in the long term. There will be times when it makes sense to use leverage, but only when future volatility is predicted to be below its average. Moving forward, I am not planning on increasing my leveraged positions. I'm in the process of building a GARCH model to predict future volatility (still researching accuracy of how many days in the future GARCH modeling can forecast with statistical significance).

I have also been working with GARCH to try to create a leveraged etf strategy. If you want to collaborate or exchange ideas please message me (I am unable to message since my account is new)
taojaxx
Posts: 128
Joined: Wed Jul 18, 2012 8:25 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by taojaxx »

tomphilly wrote: Wed Aug 05, 2020 12:15 pm Hi all, I'm new here and I have been following this thread and other discussions like it elsewhere with great interest for a while. I've been employing a UPRO/TMF 60/40 ratio since mid 2019 with good success. I share the concern of some here about interest rates and their effect on TMF in the long term. But I also acknowledge that there's a strong chance they won't rise any time soon and historically interest rates have been dropping for centuries - however, to be practical, this strategy needs to be effective now and during this decade, not during the next decade or when we look back over the past century of investment strategies on boggleheads.org in the year 2100. Clearly no one would hang on to this strategy over certain decades in the last century (60's to 80's) and the 2020's could become one of those decades.

When I think about it I have more certainty that the SP500 will rise than LTT's yields will drop over the next decade. So the result of this thought experiment is to ask what can be done to feel more comfortable with the volatility hedge. A possible rise in interest rates is just a known unknown that will affect TMF. My concern is there are unknown unknowns that could cause LTT's to deteriorate:

1) During the March crash the volatility hedge started to break down on several trading days, leading to some eye-watering drawdowns for me - I don't fully understand why, but I believe the aggressive fed intervention played some part in this strategy not being completely wiped out.
2) Today, TMF is down 3% in response to a Fed debt sale - is this a once off, or will it continue?
3) What happens to LTT's if the US dollar is no longer the world's reserve currency?
4) What happens to LTT's if we returned to the gold standard? (Not a joke - it is a policy position of the new fed candidate)

I know you might think, "Well, we can dump or adjust this strategy before these things take place", but that seems a lot like throwing a frog in boiling water (frog jumps out & survives) versus putting a frog in cold water and heating it up (frog slowly dies).

Note that I'm not nearly as financially or macroeconomically savvy as many of you here, I'm more "financially curious" as someone else put it. With my limited savviness, I'm looking at alternative volatility hedges like BTAL, TYD and UST, to see if the strategy could still be effective with a basket like 15% TMF, 15% TYD & 10% BTAL.
2) Today, TMF is down 3% in response to a Fed debt sale - is this a once off, or will it continue?
3) What happens to LTT's if the US dollar is no longer the world's reserve currency?
4) What happens to LTT's if we returned to the gold standard? (Not a joke - it is a policy position of the new fed candidate)
If you have this on your mind, there is zero reason for you to consider this strategy.
Simple as that.
rascott
Posts: 2407
Joined: Wed Apr 15, 2015 10:53 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by rascott »

jarjarM wrote: Tue Aug 04, 2020 5:29 pm
SVT wrote: Tue Aug 04, 2020 4:34 pm
mrspock wrote: Tue Aug 04, 2020 3:24 pm
Meaty wrote: Mon Aug 03, 2020 11:12 am
HawkeyePierce wrote: Mon Aug 03, 2020 12:53 am I'm just about at the one year mark of diving into the EDV flavor of the Adventure. In that time, my allocation is up 49.57%.

(43/57 UPRO/EDV, quarterly rebals, entirely in Roth space)
Started a year ago this month. 55/45 UPRO TMF. I’m up 46%.
I'm doing the 50/50 version of this w/ quarterly rebalance.

I'm up similar, I've honestly lost count....but my original investment is up north of 50% now (started Feb 2019).
Likely way north of 50%. Probably well over 100%.
Very unlikely to be over 100% if there's no additional $$$ added in. Depending on the exact date one starts this, UPRO could be still underwater (after 7/2019 entry date) or just ~20% above 2/2019 price level. While TMF is ~100% return since the original adventure started, UPRO return is subpar due to the march drawdown. Volatility decay is very real with the 3x LETF.

Personally I'm up ~70%, started around March 2019.
Yep, UPRO down 25% though July, while SP500 up 2%. That's some wicked volatility decay.
rascott
Posts: 2407
Joined: Wed Apr 15, 2015 10:53 am

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by rascott »

So I've now in my 4thv quarter of running this similar strategy using a TDA futures account. Target is 50/50 stocks/ bonds, 3x leveraged. Treasury bonds with similar total duration as TMF. Stocks are a combo of VOO + micro emini futures. Bonds are all Treasury futures.

Up approx 60% since late Aug/ early Sept. Don't have the exact starting date in front of me. But seems to have tracked the LETF strategy really closely.

I have a lot of cash built up from the daily settlements, so next month of the roll, I'll likely buy more VOO and drop an emini contract or two. It takes me about 20 mins every qtr to do the calculations of what I need and then to roll the contracts.
Veritas2
Posts: 5
Joined: Mon Apr 27, 2020 7:11 pm

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by Veritas2 »

Does anyone have more than 50% of their original portfolio in this journey?
User avatar
cos
Posts: 288
Joined: Fri Aug 23, 2019 7:34 pm
Location: Boston
Contact:

Re: HEDGEFUNDIE's excellent adventure Part II: The next journey

Post by cos »

Veritas2 wrote: Wed Aug 05, 2020 9:23 pm Does anyone have more than 50% of their original portfolio in this journey?
Yep! I'm sitting pretty with ~95% of my net worth in this strategy.
Post Reply