HEDGEFUNDIE's excellent adventure Part II: The next journey
-
- Posts: 7
- Joined: Mon Jul 05, 2021 4:16 am
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I'm a 25 year old with a long-term investment horizon and am in on the underlying rationale of this strategy.
I'm looking for some thoughts/criticisms on a couple of proposed modifications given the current state of the market:
Holding 10% in Gold: I ran backtests from 03-07' and from 15-19' in (slowly) rising interest rate environments (what I predict will be the next 5-10 years), and the funds with an allocation shifted from TMF to UGL performed better. (Am waiting on more info from GDLX before shifting to 3X - but given the inherent volatility of gold am hesitant to do this anyways).
Holding 12% in Short Term Treasuries or Cash: Yes, I know, it's impossible to time the market. However, I do believe things as they currently stand are slightly overvalued, and we are due for at a minimum a moderate pullback. Given this, my thinking is to hold a portion of my portfolio in short term treasuries (or cash) and to:
- Shift 4% into UPRO/TQQQ upon the occurrence of a 15% pullback from 52 week high of SPY/QQQ;
- Shift the last 8% into UPRO/TQQQ upon the occurrence of a 25+% pullback from 52 week high of SPY/QQQ;
- While held, short term treasuries also help to implement the barbell strategy in buying both short and long term treasuries as a hedge against rising
interest rates.
-(This bullet is difficult to backtest on a longer-term horizon, so I really have no idea what the historical effects would have looked like.
My guess would be it is entirely dependant upon how long your holding the STT/cash before finally entering the market, but getting in either of these ETF's at a favourable price seems that it would pay off.)
Holding 20% in TQQQ: This is a personal preference, really, but I'm a believer in big tech and do continue to see it outperforming the market in the next 5-10 years. The shift to 52.5/47.5 (opposed to 55-45) is simply to account for the increased volatility of TQQQ compared to the broader UPRO.
Given the above, my proposed portfolio looks like the following:
32.5% UPRO
20% TQQQ
25.5% TMF
12% VGSH/Cash
10% UGL
With quarterly rebalancing, and 12% VGSH/Cash subject to shift into UPRO/TQQQ (65/35 portfolio allocation) depending what happens in the market!
Please let me know any thoughts or criticisms.
I'm looking for some thoughts/criticisms on a couple of proposed modifications given the current state of the market:
Holding 10% in Gold: I ran backtests from 03-07' and from 15-19' in (slowly) rising interest rate environments (what I predict will be the next 5-10 years), and the funds with an allocation shifted from TMF to UGL performed better. (Am waiting on more info from GDLX before shifting to 3X - but given the inherent volatility of gold am hesitant to do this anyways).
Holding 12% in Short Term Treasuries or Cash: Yes, I know, it's impossible to time the market. However, I do believe things as they currently stand are slightly overvalued, and we are due for at a minimum a moderate pullback. Given this, my thinking is to hold a portion of my portfolio in short term treasuries (or cash) and to:
- Shift 4% into UPRO/TQQQ upon the occurrence of a 15% pullback from 52 week high of SPY/QQQ;
- Shift the last 8% into UPRO/TQQQ upon the occurrence of a 25+% pullback from 52 week high of SPY/QQQ;
- While held, short term treasuries also help to implement the barbell strategy in buying both short and long term treasuries as a hedge against rising
interest rates.
-(This bullet is difficult to backtest on a longer-term horizon, so I really have no idea what the historical effects would have looked like.
My guess would be it is entirely dependant upon how long your holding the STT/cash before finally entering the market, but getting in either of these ETF's at a favourable price seems that it would pay off.)
Holding 20% in TQQQ: This is a personal preference, really, but I'm a believer in big tech and do continue to see it outperforming the market in the next 5-10 years. The shift to 52.5/47.5 (opposed to 55-45) is simply to account for the increased volatility of TQQQ compared to the broader UPRO.
Given the above, my proposed portfolio looks like the following:
32.5% UPRO
20% TQQQ
25.5% TMF
12% VGSH/Cash
10% UGL
With quarterly rebalancing, and 12% VGSH/Cash subject to shift into UPRO/TQQQ (65/35 portfolio allocation) depending what happens in the market!
Please let me know any thoughts or criticisms.
-
- Posts: 386
- Joined: Sun Feb 04, 2018 6:44 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I have no idea what the market will do -pull back, melt up, sideways- nor do I really care: I intend to live through it. That being said, there's nothing egregious with your portfolio composition, it should survive stormy weather.Leafsfan1996 wrote: ↑Mon Jul 05, 2021 4:29 am I'm a 25 year old with a long-term investment horizon and am in on the underlying rationale of this strategy.
I'm looking for some thoughts/criticisms on a couple of proposed modifications given the current state of the market:
Holding 10% in Gold: I ran backtests from 03-07' and from 15-19' in (slowly) rising interest rate environments (what I predict will be the next 5-10 years), and the funds with an allocation shifted from TMF to UGL performed better. (Am waiting on more info from GDLX before shifting to 3X - but given the inherent volatility of gold am hesitant to do this anyways).
Holding 12% in Short Term Treasuries or Cash: Yes, I know, it's impossible to time the market. However, I do believe things as they currently stand are slightly overvalued, and we are due for at a minimum a moderate pullback. Given this, my thinking is to hold a portion of my portfolio in short term treasuries (or cash) and to:
- Shift 4% into UPRO/TQQQ upon the occurrence of a 15% pullback from 52 week high of SPY/QQQ;
- Shift the last 8% into UPRO/TQQQ upon the occurrence of a 25+% pullback from 52 week high of SPY/QQQ;
- While held, short term treasuries also help to implement the barbell strategy in buying both short and long term treasuries as a hedge against rising
interest rates.
-(This bullet is difficult to backtest on a longer-term horizon, so I really have no idea what the historical effects would have looked like.
My guess would be it is entirely dependant upon how long your holding the STT/cash before finally entering the market, but getting in either of these ETF's at a favourable price seems that it would pay off.)
Holding 20% in TQQQ: This is a personal preference, really, but I'm a believer in big tech and do continue to see it outperforming the market in the next 5-10 years. The shift to 52.5/47.5 (opposed to 55-45) is simply to account for the increased volatility of TQQQ compared to the broader UPRO.
Given the above, my proposed portfolio looks like the following:
32.5% UPRO
20% TQQQ
25.5% TMF
12% VGSH/Cash
10% UGL
With quarterly rebalancing, and 12% VGSH/Cash subject to shift into UPRO/TQQQ (65/35 portfolio allocation) depending what happens in the market!
Please let me know any thoughts or criticisms.
Only remark is I don't understand why (other than performance chasing) people complement UPRO with TQQQ whereas Tech is already a large part of UPRO. Kind of deliberately negating the benefits of diversification. So basically sending back the only free lunch.
Better lucky than smart.
-
- Posts: 7
- Joined: Mon Jul 05, 2021 4:16 am
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Thanks for the feedback. I did find the backtest linked below interesting - portfolio 1 is the general HEFA strategy compared with 1/2 UPRO and 1/2 Nasdaq since 1994. I realize it's not entirely accurate because of the limitations of the LETFS themselves only existing since 2010~, but interesting nonetheless.I have no idea what the market will do -pull back, melt up, sideways- nor do I really care: I intend to live through it. That being said, there's nothing egregious with your portfolio composition, it should survive stormy weather.
Only remark is I don't understand why (other than performance chasing) people complement UPRO with TQQQ whereas Tech is already a large part of UPRO. Kind of deliberately negating the benefits of diversification. So basically sending back the only free lunch.
https://www.portfoliovisualizer.com/bac ... l5=%5EGOLD
UPRO has a slightly higher (0.04) higher Sharpe/Sortino, but a near 3% lower CAGR. I don't foresee 1999 happening to the nasdaq again, and with that significant drop removed, I presume that these differences would only be exacerbated.
- Tinkerer-in-Chief
- Posts: 85
- Joined: Mon May 25, 2020 4:20 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Yes, I too seem to have misplaced my crystal ball.
This space intentionally left blank.
-
- Posts: 7
- Joined: Mon Jul 05, 2021 4:16 am
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Lol. 1999 was the result of the internet boom, something akin to what cryptocurrency currently is - rampant speculation with no underlying business expansion fueling future growth. I, personally, would feel much stronger betting that a 1999 event (a 70+% drop) would happen to the 2021 crypto market than the nasdaq index given the underlying nature of what is fueling the growth (speculation opposed to business expansion).Yes, I too seem to have misplaced my crystal ball.
Amazon and Apple, for example, are two of the primary holdings of the Nasdaq making up about 18%, which combined for astronomical $200 billion in revenue in Q1 2021. Surely 1999 is not impossible, but I feel much better betting that 1999 (a 70+% drop) will not happen to a Nasdaq made up of companies like this than I would have in 1999. Barring some massive change to the way businesses operate, I would say that bet holds true.
Goes without saying that any market is subject to crashes and it is impossible to determine with certainty what the future holds. This considered, there is room to instill some logic and common sense. The mere inability to predict with certainty what will happen in the future does not imply that you must ignore any and all differences between the 1999 situation and now.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I tried my risk-budget minimum variance approach with these funds, leaving out the cash part (I don't have a good way to balance volatility between high- and low-volatility assets). You could think of this as 88 percent of the portfolio.Leafsfan1996 wrote: ↑Mon Jul 05, 2021 9:22 amThanks for the feedback. I did find the backtest linked below interesting - portfolio 1 is the general HEFA strategy compared with 1/2 UPRO and 1/2 Nasdaq since 1994. I realize it's not entirely accurate because of the limitations of the LETFS themselves only existing since 2010~, but interesting nonetheless.Only remark is I don't understand why (other than performance chasing) people complement UPRO with TQQQ whereas Tech is already a large part of UPRO. Kind of deliberately negating the benefits of diversification. So basically sending back the only free lunch.
That's like saying using UPRO complemented with SPXL negates diversification, because SPXL is already a large part of UPRO...
https://www.portfoliovisualizer.com/bac ... l5=%5EGOLD
UPRO has a slightly higher (0.04) higher Sharpe/Sortino, but a near 3% lower CAGR. I don't foresee 1999 happening to the nasdaq again, and with that significant drop removed, I presume that these differences would only be exacerbated.
Assuming that equities have 5 and 30 times the risk budgeted for LTT and gold, respectively, and UPRO and TQQQ each have the half of the equities risk budget, and rebalancing every two weeks, gives you something like
The funds are synthetic prior to inception, and quite likely optimistic.
So over the long run UGL is looking like more of a store of value than a risk balance for this portfolio, even though it is 2x.
This would be about 23/16/12/37/12 average allocation to UPRO/TQQQ/UGL/TMF/VGSH.
To get the proportion you are targeting, basically the equities risk budget would have been ~10 and ~50 times the risk budget for LTT and gold. I get an increase in CAGR by ~9 percent and decrease in Sharpe by ~9 percent moving to the new allocation.
You won't get the same performance with quarterly rebalancing and a fixed allocation, it'll likely be quite a bit bumpier, but at least this gives you some independent look at the interplay between risk and allocation.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Agreed, and one of the major differences that people seem to forget is that at the time interest rates were at 6%. Now we are at 0.25%. P/E ratios are not even close to the same when you take this into account.Leafsfan1996 wrote: ↑Mon Jul 05, 2021 10:15 amLol. 1999 was the result of the internet boom, something akin to what cryptocurrency currently is - rampant speculation with no underlying business expansion fueling future growth. I, personally, would feel much stronger betting that a 1999 event (a 70+% drop) would happen to the 2021 crypto market than the nasdaq index given the underlying nature of what is fueling the growth (speculation opposed to business expansion).Yes, I too seem to have misplaced my crystal ball.
Amazon and Apple, for example, are two of the primary holdings of the Nasdaq making up about 18%, which combined for astronomical $200 billion in revenue in Q1 2021. Surely 1999 is not impossible, but I feel much better betting that 1999 (a 70+% drop) will not happen to a Nasdaq made up of companies like this than I would have in 1999. Barring some massive change to the way businesses operate, I would say that bet holds true.
Goes without saying that any market is subject to crashes and it is impossible to determine with certainty what the future holds. This considered, there is room to instill some logic and common sense. The mere inability to predict with certainty what will happen in the future does not imply that you must ignore any and all differences between the 1999 situation and now.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
One year.
-
- Posts: 7
- Joined: Mon Jul 05, 2021 4:16 am
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Agreed, and one of the major differences that people seem to forget is that at the time interest rates were at 6%. Now we are at 0.25%. P/E ratios are not even close to the same when you take this into account.
Completely agree. Reoccurrence is surely possible - but basing that conclusion of a comparison of the 1999 market to now it is comparing apples to oranges.
-
- Posts: 7
- Joined: Mon Jul 05, 2021 4:16 am
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Very interesting. Thanks for running this!The funds are synthetic prior to inception, and quite likely optimistic.
So over the long run UGL is looking like more of a store of value than a risk balance for this portfolio, even though it is 2x.
This would be about 23/16/12/37/12 average allocation to UPRO/TQQQ/UGL/TMF/VGSH.
To get the proportion you are targeting, basically the equities risk budget would have been ~10 and ~50 times the risk budget for LTT and gold. I get an increase in CAGR by ~9 percent and decrease in Sharpe by ~9 percent moving to the new allocation.
You won't get the same performance with quarterly rebalancing and a fixed allocation, it'll likely be quite a bit bumpier, but at least this gives you some independent look at the interplay between risk and allocation.
-
- Posts: 158
- Joined: Thu Feb 25, 2021 11:43 am
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
So I noticed a lot of people hating on HFEA because of volatility drag and the costs potentially going up, but the cost of any form of leverage goes up if rates go up too.
Interest rates up, margin rates go up, implied borrowing cost of LEAPS up.
Also, I don't get the fascination of LEAPs. They're tax inefficient, come with additional calculations and varying levels of leverage as the time changes, have bid ask spreads that can get ugly sometimes, I calculate the costs of LEAPs, with small amounts can be a lot but with more money and tight bid asks spread should still be within a rounding error of LEAPs. The tax inefficiency kills them though.
Margin with IBKR is great but the risk of wipeout and automatic sell offs sucks. You also can't acquire as much leverage. Forced liquidation is a real risk.
Do LETFs get an unfair rep?
The CAGR of using margin with SPY to 2x vs SSO results in 2% lower CAGR for SSO. However, a lump sump in SSO has the tax efficiency, simplicity, doesn't require you to make any decisions.
Ultimately, each method, I think, is within a few bps of the other in the end I suspect.
Interest rates up, margin rates go up, implied borrowing cost of LEAPS up.
Also, I don't get the fascination of LEAPs. They're tax inefficient, come with additional calculations and varying levels of leverage as the time changes, have bid ask spreads that can get ugly sometimes, I calculate the costs of LEAPs, with small amounts can be a lot but with more money and tight bid asks spread should still be within a rounding error of LEAPs. The tax inefficiency kills them though.
Margin with IBKR is great but the risk of wipeout and automatic sell offs sucks. You also can't acquire as much leverage. Forced liquidation is a real risk.
Do LETFs get an unfair rep?
The CAGR of using margin with SPY to 2x vs SSO results in 2% lower CAGR for SSO. However, a lump sump in SSO has the tax efficiency, simplicity, doesn't require you to make any decisions.
Ultimately, each method, I think, is within a few bps of the other in the end I suspect.
-
- Posts: 7
- Joined: Mon Jul 05, 2021 4:16 am
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Is it possible to DCA this without wrecking the quarterly rebalancing?
For example - if I get paid a bi weekly pay check and want to deposit every 2 weeks. Would I deposit at the current ratios, and then only adjust the %s on a quarterly basis. For example, if I was implementing strict HEFA 55-45 with quarterly rebalanced, and at the time of my pay check, the ratio was 62-38 - would I deposit at those ratios? Or would I deposit to bring it back in line with its 55-45 target?
For example - if I get paid a bi weekly pay check and want to deposit every 2 weeks. Would I deposit at the current ratios, and then only adjust the %s on a quarterly basis. For example, if I was implementing strict HEFA 55-45 with quarterly rebalanced, and at the time of my pay check, the ratio was 62-38 - would I deposit at those ratios? Or would I deposit to bring it back in line with its 55-45 target?
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Just contribute quarterly instead of every two weeks if you want to stay true. The point of the portfolio is to let the winners run and the losers suck wind for the quarter. If you want to contribute every 2 weeks, contribute at the then split (i.e. if you’re 62/38, contribute at 62/38).Leafsfan1996 wrote: ↑Tue Jul 06, 2021 9:28 am Is it possible to DCA this without wrecking the quarterly rebalancing?
For example - if I get paid a bi weekly pay check and want to deposit every 2 weeks. Would I deposit at the current ratios, and then only adjust the %s on a quarterly basis. For example, if I was implementing strict HEFA 55-45 with quarterly rebalanced, and at the time of my pay check, the ratio was 62-38 - would I deposit at those ratios? Or would I deposit to bring it back in line with its 55-45 target?
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Due to the bucketing approach most are taking to this adventure, the general consensus is to run this primarily inside a Roth. Since you cannot use portfolio margin in retirement accounts, that leaves only LETFs or LEAPS, with taxes basically ignored.TheDoctor91 wrote: ↑Tue Jul 06, 2021 12:21 am So I noticed a lot of people hating on HFEA because of volatility drag and the costs potentially going up, but the cost of any form of leverage goes up if rates go up too.
Interest rates up, margin rates go up, implied borrowing cost of LEAPS up.
Also, I don't get the fascination of LEAPs. They're tax inefficient, come with additional calculations and varying levels of leverage as the time changes, have bid ask spreads that can get ugly sometimes, I calculate the costs of LEAPs, with small amounts can be a lot but with more money and tight bid asks spread should still be within a rounding error of LEAPs. The tax inefficiency kills them though.
Margin with IBKR is great but the risk of wipeout and automatic sell offs sucks. You also can't acquire as much leverage. Forced liquidation is a real risk.
Do LETFs get an unfair rep?
The CAGR of using margin with SPY to 2x vs SSO results in 2% lower CAGR for SSO. However, a lump sump in SSO has the tax efficiency, simplicity, doesn't require you to make any decisions.
Ultimately, each method, I think, is within a few bps of the other in the end I suspect.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I would deposit to bring back to your target if it is convenient, but have no heartburn at just depositing at the desired ratio if it isn't. Depositing it to bring back to the target would eliminate rebalancing at first, but after a while you won't be able to avoid rebalancing.Leafsfan1996 wrote: ↑Tue Jul 06, 2021 9:28 am Is it possible to DCA this without wrecking the quarterly rebalancing?
For example - if I get paid a bi weekly pay check and want to deposit every 2 weeks. Would I deposit at the current ratios, and then only adjust the %s on a quarterly basis. For example, if I was implementing strict HEFA 55-45 with quarterly rebalanced, and at the time of my pay check, the ratio was 62-38 - would I deposit at those ratios? Or would I deposit to bring it back in line with its 55-45 target?
M1 automatically sets it to bring back to the target, which is nice.
-
- Posts: 1594
- Joined: Mon Aug 24, 2020 5:44 pm
-
- Posts: 2344
- Joined: Tue Mar 05, 2019 9:29 pm
- Location: Colorado
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I rebalanced today after forgetting to do it on the 1st, though my allocations were only off from my target by about 2%.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I'm still in, at 60:40 also. I sold a small amount of UPRO @ $112.75 and bought TMF @ $26.42, reducing TMF cost basis a tiny amount from $30.30 to $30.15.
/FIRE55
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I'm sooo close to getting back to breakeven on TMF. Was all the way up above $40 cost basis at one point, bought a whole bunch at $22.80 which ended up being an excellent choice, now my cost basis is down to $28.40.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Nice. I had doubled my money in TMF by July 2020 (cb $22.33, price $44.56). Kept rebalancing, cost basis peaked at $34.08, now down to $30.15. On March 18th I bought a batch of TMF at $21.13, those are doing well. Fascinating.
/FIRE55
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Was down on TMF over 30% now only 7%.
Get rich or die tryin'
-
- Posts: 1
- Joined: Tue Jul 06, 2021 4:17 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Hi everyone,
I've been participating in this excellent adventure for some time, but its my first time registering and posting.
I have a HFEA thought experiment that I hope that you can help me out with:
We know that a simple Portfolio Visualizer back-test of 55% UPRO, 45% TMF, with quarterly rebalancing gives a GAGR of 34.83% (from Jan 2010 to Jun 2021).
We also know that if you drop the rebalancing completely, the CAGR drops dramatically, in this case to 27.96%.
I have half my HFEA in a tax-sheltered account, and half in a taxable account.
So to avoid triggering taxes when I rebalance every quarter, I am only rebalancing the tax-sheltered account; but when I rebalance, I still factor-in the entire portfolio (taxable and non-taxable) holdings.
For example, if my target portfolio is 55% UPRO and 45% TMF, and if right before rebalancing, the accounts are:
Taxable:
- UPRO: 50%
- TMF: 50%
Tax-sheltered:
- UPRO: 50%
- TMF: 50%
Then after rebalancing, the balances will be:
Taxable:
- UPRO: 50% <- no rebalancing
- TMF: 50% <- no rebalancing
Tax-sheltered:
- UPRO: 60% <- over-compensates for taxable account that won't be rebalanced
- TMF: 40% <- over-compensates for taxable account that won't be rebalanced
So that the overall portfolio after rebalancing will still maintain the target of:
- UPRO: 55%
- TMF: 45%
In theory, the above method should still give the same returns (GAGR of 34.83%) as if my entire portfolio was rebalanced.
So where the thought experiment comes in is as follows:
Given that:
A. Rebalancing an entire portfolio quarterly gave 34.83% CAGR
B. No rebalancing entire portfolio quarterly gave 27.96% CAGR (this is equivalent to my taxable account)
C. My portfolio is comprised half of a no-rebalancing (i.e. 27.96% CAGR) and half of over-compensation rebalancing.
Then wouldn't this mean that my tax-sheltered account is achieving a 41.7% CAGR in order that my overall portfolio still achieves the 34.83% CAGR average?
And if so, then does that mean a more effective means of rebalancing our HFEA portfolios is to over-rebalance ; where the general rule would be:
Target Portfolio (e.g.):
- UPRO: 55%
- TMF: 45%
Portfolio Before Rebalancing:
- UPRO: 55% + X% (e.g. 55% + 10% = 65%)
- TMF: 45% - X% (e.g. 45% - 10% = 35%)
Portfolio After Rebalancing:
- UPRO: 55% - X% (i.e. 55% - 10% = 45%) <- as opposed to 55%
- TMF: 45% + X% (i.e. 55% + 10% = 55%) <- as opposed to 45%
If some of you superstars could chime-in, it would be appreciated.
Thank you.
I've been participating in this excellent adventure for some time, but its my first time registering and posting.
I have a HFEA thought experiment that I hope that you can help me out with:
We know that a simple Portfolio Visualizer back-test of 55% UPRO, 45% TMF, with quarterly rebalancing gives a GAGR of 34.83% (from Jan 2010 to Jun 2021).
We also know that if you drop the rebalancing completely, the CAGR drops dramatically, in this case to 27.96%.
I have half my HFEA in a tax-sheltered account, and half in a taxable account.
So to avoid triggering taxes when I rebalance every quarter, I am only rebalancing the tax-sheltered account; but when I rebalance, I still factor-in the entire portfolio (taxable and non-taxable) holdings.
For example, if my target portfolio is 55% UPRO and 45% TMF, and if right before rebalancing, the accounts are:
Taxable:
- UPRO: 50%
- TMF: 50%
Tax-sheltered:
- UPRO: 50%
- TMF: 50%
Then after rebalancing, the balances will be:
Taxable:
- UPRO: 50% <- no rebalancing
- TMF: 50% <- no rebalancing
Tax-sheltered:
- UPRO: 60% <- over-compensates for taxable account that won't be rebalanced
- TMF: 40% <- over-compensates for taxable account that won't be rebalanced
So that the overall portfolio after rebalancing will still maintain the target of:
- UPRO: 55%
- TMF: 45%
In theory, the above method should still give the same returns (GAGR of 34.83%) as if my entire portfolio was rebalanced.
So where the thought experiment comes in is as follows:
Given that:
A. Rebalancing an entire portfolio quarterly gave 34.83% CAGR
B. No rebalancing entire portfolio quarterly gave 27.96% CAGR (this is equivalent to my taxable account)
C. My portfolio is comprised half of a no-rebalancing (i.e. 27.96% CAGR) and half of over-compensation rebalancing.
Then wouldn't this mean that my tax-sheltered account is achieving a 41.7% CAGR in order that my overall portfolio still achieves the 34.83% CAGR average?
And if so, then does that mean a more effective means of rebalancing our HFEA portfolios is to over-rebalance ; where the general rule would be:
Target Portfolio (e.g.):
- UPRO: 55%
- TMF: 45%
Portfolio Before Rebalancing:
- UPRO: 55% + X% (e.g. 55% + 10% = 65%)
- TMF: 45% - X% (e.g. 45% - 10% = 35%)
Portfolio After Rebalancing:
- UPRO: 55% - X% (i.e. 55% - 10% = 45%) <- as opposed to 55%
- TMF: 45% + X% (i.e. 55% + 10% = 55%) <- as opposed to 45%
If some of you superstars could chime-in, it would be appreciated.
Thank you.
-
- Posts: 7
- Joined: Mon Jul 05, 2021 4:16 am
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I could be wrong, and this is simply based off my initial presumption without running any analysis or backtests. However, this seems to false presume that the portfolio that is benefitting you will be the over-balanced portfolio - ie. that rebalancing 60-40 to UPRO would always be favourable and thereby give a higher return. In reality, however, your 50/50 portfolio would outperform the 60-40 one in a bear market. There is a disconnect between the 2010-2021 sampling period that you are using, and the 1955-2021 period in which 55-45 is based off of.Then wouldn't this mean that my tax-sheltered account is achieving a 41.7% CAGR in order that my overall portfolio still achieves the 34.83% CAGR average?
The 34% CAGR you're speaking of since 2010 is conveniently commencing right at the beginning of one of the longest bull runs in history. The 34% is also, importantly, an aggregate of your entirely portfolios returns. So, for example, if the market went down 20%, your overbalanced portfolio would then do significantly worse than the one that remained unchanged. If you ran a hypothetical backtest on how the portfolio would perform during a bear market or a crash, the other would outperform. If conditions since 2010 were present since 1955, your overbalanced portfolio would be the best pick (well, actually, a maybe something closer to 75-25 - https://www.portfoliovisualizer.com/bac ... tion2_3=25 ). However, the 55-45 ratio is what has performed optimally from 1955-2021, 1985-2021, etc. (much longer periods of time with significant bear markets).
The point of HEFA is that it is balanced precisely in such a way as to succeed to the greatest extent possible without being subject to a pullback that would make recovery near impossible (such as if you held 100% UPRO during 08'). While greater returns are possible - nothing has done better over the longer term.
I, too, have my HEFA portfolio across multiple accounts. I have found that having a portfolio tracker with its components listed in aggregate amounts to be extremely helpful for this very issue.
Others - please chime in if I am missing something.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I suggest that you perform a calculation with real numbers for each of the portfolio assets. I suspect that you will find that rebalancing won't be as straightforward as it seems at first glance.
Before the first rebalance, the tax-advantaged and taxable accounts grow at the same rate.
After the first rebalance, the two accounts grow at different rates. This may end up making the tax-advantaged rebalances larger and larger over time to compensate for the increasingly out of balance taxable account. The two accounts will grow at increasingly different rates, and at some point it won't be possible to rebalance.
Before the first rebalance, the tax-advantaged and taxable accounts grow at the same rate.
After the first rebalance, the two accounts grow at different rates. This may end up making the tax-advantaged rebalances larger and larger over time to compensate for the increasingly out of balance taxable account. The two accounts will grow at increasingly different rates, and at some point it won't be possible to rebalance.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Give me 30% CAGR and you can increase my tax bracket to the next one.
Get rich or die tryin'
-
- Posts: 445
- Joined: Mon Mar 23, 2020 2:03 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Deleted
Last edited by fatcoffeedrinker on Wed Mar 02, 2022 10:08 am, edited 1 time in total.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
fatcoffeedrinker wrote: ↑Wed Jul 07, 2021 11:05 am Just thought I would add a data point. I got into HFEA in February, 2019 with $100K. Started with original 40/60, then switched to 55/45 when Part II came out. Have never added more -- just quarterly rebalancing. Just hit $300K today! Woohoo!
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Awesome. CAGR of ~56%!fatcoffeedrinker wrote: ↑Wed Jul 07, 2021 11:05 am Just thought I would add a data point. I got into HFEA in February, 2019 with $100K. Started with original 40/60, then switched to 55/45 when Part II came out. Have never added more -- just quarterly rebalancing. Just hit $300K today! Woohoo!
/FIRE55
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Phenomenal! Congrats!fatcoffeedrinker wrote: ↑Wed Jul 07, 2021 11:05 am Just thought I would add a data point. I got into HFEA in February, 2019 with $100K. Started with original 40/60, then switched to 55/45 when Part II came out. Have never added more -- just quarterly rebalancing. Just hit $300K today! Woohoo!
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Congratulations. What is your exit strategy. Are you holding for a certain period of time or until you reach a particular dollar amount?fatcoffeedrinker wrote: ↑Wed Jul 07, 2021 11:05 am Just thought I would add a data point. I got into HFEA in February, 2019 with $100K. Started with original 40/60, then switched to 55/45 when Part II came out. Have never added more -- just quarterly rebalancing. Just hit $300K today! Woohoo!
-
- Posts: 445
- Joined: Mon Mar 23, 2020 2:03 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Deleted
Last edited by fatcoffeedrinker on Wed Mar 02, 2022 10:08 am, edited 1 time in total.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I'm also tending towards something like a 10% allocation (i.e. 6% UPRO 4% TMF) within a (broadly) 60:40 allocated portfolio.fatcoffeedrinker wrote: ↑Wed Jul 07, 2021 12:09 pmI don't know yet, but I do need to figure that out. When I first started, it was 4.1% of our overall portfolio. It is now 6.5% of our portfolio (our non-HFEA portfolio is up 85% since then, including contributions) and less than 5% of overall NW. So it is still not huge on a relative basis, and if it went to zero (the probability of which I think is nil), it would be a really bad day, but not life-changing. I'm thinking if/when it breaks 10% of the portfolio, I'll start to sell down to keep it from getting too much larger than 10%.Ramjet wrote: ↑Wed Jul 07, 2021 11:42 amCongratulations. What is your exit strategy. Are you holding for a certain period of time or until you reach a particular dollar amount?fatcoffeedrinker wrote: ↑Wed Jul 07, 2021 11:05 am Just thought I would add a data point. I got into HFEA in February, 2019 with $100K. Started with original 40/60, then switched to 55/45 when Part II came out. Have never added more -- just quarterly rebalancing. Just hit $300K today! Woohoo!
viewtopic.php?p=6054676#p6054676
/FIRE55
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I am letting it ride. I've compartmentalized the investment the best I can and I am trying to be emotionless about it.FIRE55 wrote: ↑Wed Jul 07, 2021 12:34 pmI'm also tending towards something like a 10% allocation (i.e. 6% UPRO 4% TMF) within a (broadly) 60:40 allocated portfolio.fatcoffeedrinker wrote: ↑Wed Jul 07, 2021 12:09 pmI don't know yet, but I do need to figure that out. When I first started, it was 4.1% of our overall portfolio. It is now 6.5% of our portfolio (our non-HFEA portfolio is up 85% since then, including contributions) and less than 5% of overall NW. So it is still not huge on a relative basis, and if it went to zero (the probability of which I think is nil), it would be a really bad day, but not life-changing. I'm thinking if/when it breaks 10% of the portfolio, I'll start to sell down to keep it from getting too much larger than 10%.Ramjet wrote: ↑Wed Jul 07, 2021 11:42 amCongratulations. What is your exit strategy. Are you holding for a certain period of time or until you reach a particular dollar amount?fatcoffeedrinker wrote: ↑Wed Jul 07, 2021 11:05 am Just thought I would add a data point. I got into HFEA in February, 2019 with $100K. Started with original 40/60, then switched to 55/45 when Part II came out. Have never added more -- just quarterly rebalancing. Just hit $300K today! Woohoo!
viewtopic.php?p=6054676#p6054676
/FIRE55
-
- Posts: 717
- Joined: Wed Mar 25, 2009 12:58 am
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
You win the balls of steel award.fatcoffeedrinker wrote: ↑Wed Jul 07, 2021 11:05 am Just thought I would add a data point. I got into HFEA in February, 2019 with $100K. Started with original 40/60, then switched to 55/45 when Part II came out. Have never added more -- just quarterly rebalancing. Just hit $300K today! Woohoo!
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Congrats. I started ~1 month later at March 10th, 2019. Just saw it hitting $300k ($100k initial and only investment) as well, my rebalancing method is a bit different but result is the same.fatcoffeedrinker wrote: ↑Wed Jul 07, 2021 11:05 am Just thought I would add a data point. I got into HFEA in February, 2019 with $100K. Started with original 40/60, then switched to 55/45 when Part II came out. Have never added more -- just quarterly rebalancing. Just hit $300K today! Woohoo!
-
- Posts: 445
- Joined: Mon Mar 23, 2020 2:03 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Deleted
Last edited by fatcoffeedrinker on Wed Mar 02, 2022 10:08 am, edited 1 time in total.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
That's great!fatcoffeedrinker wrote: ↑Wed Jul 07, 2021 11:05 am Just thought I would add a data point. I got into HFEA in February, 2019 with $100K. Started with original 40/60, then switched to 55/45 when Part II came out. Have never added more -- just quarterly rebalancing. Just hit $300K today! Woohoo!
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I started June 2019 (missed some of the 2019 spring runup). I switched to TQQQ/TMF after a while. I got greedy in spring 2020, went SPXU for 3 weeks starting just at the bottom.
Let's just say that has been an increasingly expensive decision. I keep track of my poor decisions as a reminder. If I'd held steady simply holding my 2/2020 allocation, I'd be up 318 percent overall. I'm only at 60% of that portfolio and being left in the dust (a great illustration of the power of 3x compounding during bull runs!).
But I think that I've developed a much sounder and more disciplined approach for the duration, so it's all for the best to get the wild oats sown early.
Let's just say that has been an increasingly expensive decision. I keep track of my poor decisions as a reminder. If I'd held steady simply holding my 2/2020 allocation, I'd be up 318 percent overall. I'm only at 60% of that portfolio and being left in the dust (a great illustration of the power of 3x compounding during bull runs!).
But I think that I've developed a much sounder and more disciplined approach for the duration, so it's all for the best to get the wild oats sown early.
-
- Posts: 445
- Joined: Mon Mar 23, 2020 2:03 pm
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Deleted
Last edited by fatcoffeedrinker on Wed Mar 02, 2022 10:08 am, edited 1 time in total.
-
- Posts: 1
- Joined: Tue Jul 06, 2021 9:21 am
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Lurking on the forum I see that you do a lot of refinements to the hefa approach. I'm thinking about switching to adaptive asset allocation minvar strategy with 1 months timings and volatility periods for [tmf;tqqq;upro] holding 2 asset at a time. What do you think about this approach?Hydromod wrote: ↑Wed Jul 07, 2021 2:30 pm I started June 2019 (missed some of the 2019 spring runup). I switched to TQQQ/TMF after a while. I got greedy in spring 2020, went SPXU for 3 weeks starting just at the bottom.
Let's just say that has been an increasingly expensive decision. I keep track of my poor decisions as a reminder. If I'd held steady simply holding my 2/2020 allocation, I'd be up 318 percent overall. I'm only at 60% of that portfolio and being left in the dust (a great illustration of the power of 3x compounding during bull runs!).
But I think that I've developed a much sounder and more disciplined approach for the duration, so it's all for the best to get the wild oats sown early.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Been lurking, running HFEA for a short period so far, does anyone have a rebalance plan quarterly but also an upper and lower bound that triggers a rebalance? For example if UPRO drops to 40% you would rebalance.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I have tried playing with a few timing models but I haven't ever fixed on something that I was comfortable with. So far my backtesting has almost always gotten tripped up by whipsaws and such when I tried testing all-in/all-out strategies. I find that the more successful ones have stopped working as well in the last five or ten years, perhaps because the market is reacting faster with all the automated trading strategies. I can't convince myself that performance isn't mostly due to timing luck.rovidfaszu wrote: ↑Wed Jul 07, 2021 7:08 pm Lurking on the forum I see that you do a lot of refinements to the hefa approach. I'm thinking about switching to adaptive asset allocation minvar strategy with 1 months timings and volatility periods for [tmf;tqqq;upro] holding 2 asset at a time. What do you think about this approach?
One way to generate a fair comparison is to calculate the average allocation for each asset and compare the TAA approach to the fixed allocation. It seems that I usually end up worse off with the TAA portfolio. With that said, you don't know what the average allocation will be ahead of time, so there's something to be said for calculating on the fly.
I did a quick PV backtest with VFINX/RYOCX/VUSTX, holding 2 out of 3 with min variance and monthly rebalancing, and that did do better than an equal weight. However, the outperformance was basically from one month in 2008; switching to weekly rebalancing ended up with minimal performance difference, suggesting that luck played a big role.
I did better giving the algorithm a larger set of options to select 2 or 3 from, roughly the best half. I'd recommend including options like DRN, UTSL, FAS, and maybe some of the international 3x LETFs to your selection list.
*****
You might check out the work by StrategyDriven on Global Navigator and The Russell. These are dual momentum strategies. There's a variety of flavors and levels of leverage that are developed. The most relevant one is called MAX PAIN, which does one of UPRO/MIDU/URTY/TMF/EDV. It's not a smooth ride. Your proposed approach should be smoother.
*****
I currently lean towards the opposite end of allocation, because I want the smoothest ride consistent with good returns. I'm a big fan of keeping several diversified assets but using a risk budget combined with minimum variance to size the allocation. The risk budget lets me assign a fraction of risk to equities and a fraction to bonds, regardless of how many assets in each category are included (unfortunately PV doesn't have this option). I use 75/25 risk allocated to equities/bonds, which roughly corresponds to a 60/40 portfolio.
The approach tends to keep a relatively smooth growth pattern by (i) damping down allocations to volatile assets, (ii) ramping up allocations to assets that are in a low-volatility period, and (iii) taking advantage of diversification to smooth out fluctuations and get a little volatility pumping going. The approach helps improve the Sharpe ratio by reducing portfolio volatility, which reduces portfolio volatility decay, which makes compounding more effective (especially for leveraged ETFs). You explicitly set the fraction of portfolio volatility assigned to each asset (not the asset allocation), and the algorithm calculates the asset allocation. No thinking.
I'm starting to think about approaches to meet in the middle, perhaps along the lines of having a variety of assets and tossing out the current poor choices. I would use the risk budget approach, always keep the TMF component, and retain several equities, just try to switch out the poorer performers each time.
Hope this helps.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
So many alternate strategies and allocations out there. Are any conclusively better in theory or significantly better in performance than the original that Hedgefundie proposed? I haven't been convinced of thatHydromod wrote: ↑Wed Jul 07, 2021 10:07 pm I have tried playing with a few timing models but I haven't ever fixed on something that I was comfortable with. So far my backtesting has almost always gotten tripped up by whipsaws and such when I tried testing all-in/all-out strategies. I find that the more successful ones have stopped working as well in the last five or ten years, perhaps because the market is reacting faster with all the automated trading strategies. I can't convince myself that performance isn't mostly due to timing luck.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
It depends on what you mean by "significantly better."Ramjet wrote: ↑Thu Jul 08, 2021 7:46 amSo many alternate strategies and allocations out there. Are any conclusively better in theory or significantly better in performance than the original that Hedgefundie proposed? I haven't been convinced of thatHydromod wrote: ↑Wed Jul 07, 2021 10:07 pm I have tried playing with a few timing models but I haven't ever fixed on something that I was comfortable with. So far my backtesting has almost always gotten tripped up by whipsaws and such when I tried testing all-in/all-out strategies. I find that the more successful ones have stopped working as well in the last five or ten years, perhaps because the market is reacting faster with all the automated trading strategies. I can't convince myself that performance isn't mostly due to timing luck.
I would argue that you can get a smoother ride that is less affected by timing luck, perhaps with a smidge of long-term performance increase, by being attentive to risk balancing and diversification (see my example). You can increase the Sharpe ratio, if not necessarily CAGR over long durations.
That's just taking advantage of the free lunch of diversification, and is a consequence of the central limit theorem. The original HFEA does a fair amount of diversification by using a diversified index fund; it certainly can be taken further, but there's a point of diminishing returns.
If you are running HFEA as a small part of a portfolio and rebalancing, you may not get much bang for your buck because rebalancing with the unleveraged portion will dominate diversification. If it is a standalone portfolio or a large fraction of the overall portfolio, diversification will have a bigger influence.
Whether that extra effort to manage diversification is worth it really depends on the individual. It's likely to be increasingly important the closer to decumulation, when you want to mitigate sequence of returns issues.
And one of the devil's details is in picking the particular assets; the optimal assets are never known ahead of time, and you need to take some leap of faith with the collection you go ahead with.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
None to my knowledge. I tried out risk parity approach for a year with real money but then gave up, the month to month change in positions were extreme, monthly transactions were too much, and the total portfolio lagged simple 60/40 approach. I switched to 60/40 since then.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
Similar experience hereAnilG wrote: ↑Thu Jul 08, 2021 9:04 amNone to my knowledge. I tried out risk parity approach for a year with real money but then gave up, the month to month change in positions were extreme, monthly transactions were too much, and the total portfolio lagged simple 60/40 approach. I switched to 60/40 since then.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
The change in positions can be rather unattractive, especially in a taxable account. It really shouldn't be done outside of tax-advantaged.Ramjet wrote: ↑Thu Jul 08, 2021 9:09 amSimilar experience hereAnilG wrote: ↑Thu Jul 08, 2021 9:04 amNone to my knowledge. I tried out risk parity approach for a year with real money but then gave up, the month to month change in positions were extreme, monthly transactions were too much, and the total portfolio lagged simple 60/40 approach. I switched to 60/40 since then.
Going to a risk parity approach won't mirror a fixed allocation, and you have to be ok with that. It tends to flatten the peaks as well as the valleys. It's not for everyone, and that's fine.
But to be fair, a straight risk parity approach assigns half of the risk to UPRO and half to TMF, which is roughly equivalent to a 40/60 portfolio. The 60/40 is roughly equivalent to 75/25 risk weighting. It's not surprising that a portfolio taking less equity risk didn't perform as well as a portfolio taking more equity risk.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
No. I’d recommend tuning out the noise — and there’s a lot of noise in this thread.Ramjet wrote: ↑Thu Jul 08, 2021 7:46 amSo many alternate strategies and allocations out there. Are any conclusively better in theory or significantly better in performance than the original that Hedgefundie proposed? I haven't been convinced of thatHydromod wrote: ↑Wed Jul 07, 2021 10:07 pm I have tried playing with a few timing models but I haven't ever fixed on something that I was comfortable with. So far my backtesting has almost always gotten tripped up by whipsaws and such when I tried testing all-in/all-out strategies. I find that the more successful ones have stopped working as well in the last five or ten years, perhaps because the market is reacting faster with all the automated trading strategies. I can't convince myself that performance isn't mostly due to timing luck.
Re: HEDGEFUNDIE's excellent adventure Part II: The next journey
I played with doing that for a bit but ultimately found that it was hurting performance. Just rebalance quarterly, and you'll be fine.