HEDGEFUNDIE's excellent adventure Part II: The next journey

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

Hydromod wrote: Tue Aug 27, 2019 4:23 pm

Just to say that quarterly is probably about equivalent to monthly, rather than significantly better.
How about with regards to drawdown?

Annual rebalancing also seems to have merit and really is only adversely impacted in Portfolio Visualizer starting dates at-least by Black Monday.
wise_magus
Posts: 15
Joined: Sun Sep 17, 2017 1:20 pm

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

Post by wise_magus »

Hydromod wrote: Tue Aug 27, 2019 4:23 pm The calculations I've done with different rebalance periods show that daily rebalancing mitigates some of the volatility drag, which adds a percentage point or two to the CAGR. This bonus drops off rapidly as the inverse of the number of days between rebalances. I've thought that an ETF would be the best way to implement this daily rebalance.
Just to be clear: You're saying that if there were an ETF that did a daily rebalance and daily leverage reset of a 50/50 SPY/TLT portfolio, your calculations show that be historically superior to a (monthly/quarterly rebalanced) 50/50 UPRO/TMF portfolio by something like 100-200 bps? (Or a similar statement with some other choice of ratio instead of 50/50.)

If that's the case, I wonder why such a product doesn't exist. It seems like this "excellent adventure" has become quite popular and everyone would love to get an extra 100-200 bps of return.
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

wise_magus wrote: Tue Aug 27, 2019 5:34 pm
Hydromod wrote: Tue Aug 27, 2019 4:23 pm The calculations I've done with different rebalance periods show that daily rebalancing mitigates some of the volatility drag, which adds a percentage point or two to the CAGR. This bonus drops off rapidly as the inverse of the number of days between rebalances. I've thought that an ETF would be the best way to implement this daily rebalance.
Just to be clear: You're saying that if there were an ETF that did a daily rebalance and daily leverage reset of a 50/50 SPY/TLT portfolio, your calculations show that be historically superior to a (monthly/quarterly rebalanced) 50/50 UPRO/TMF portfolio by something like 100-200 bps? (Or a similar statement with some other choice of ratio instead of 50/50.)

If that's the case, I wonder why such a product doesn't exist. It seems like this "excellent adventure" has become quite popular and everyone would love to get an extra 100-200 bps of return.
Pimco StocksPLUS funds do exactly this with 100% S&P500 exposure plus various bond options.
Hydromod
Posts: 434
Joined: Tue Mar 26, 2019 10:21 pm

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

Post by Hydromod »

wise_magus wrote: Tue Aug 27, 2019 5:34 pm
Hydromod wrote: Tue Aug 27, 2019 4:23 pm The calculations I've done with different rebalance periods show that daily rebalancing mitigates some of the volatility drag, which adds a percentage point or two to the CAGR. This bonus drops off rapidly as the inverse of the number of days between rebalances. I've thought that an ETF would be the best way to implement this daily rebalance.
Just to be clear: You're saying that if there were an ETF that did a daily rebalance and daily leverage reset of a 50/50 SPY/TLT portfolio, your calculations show that be historically superior to a (monthly/quarterly rebalanced) 50/50 UPRO/TMF portfolio by something like 100-200 bps? (Or a similar statement with some other choice of ratio instead of 50/50.)

If that's the case, I wonder why such a product doesn't exist. It seems like this "excellent adventure" has become quite popular and everyone would love to get an extra 100-200 bps of return.
No, I was referring to the 50/50 UPRO/TMF (actually 40/60, but anywhere from 30/70 to 60/40 should have been similar). The daily rebalance for UPRO/TMF would have been superior to the monthly/quarterly UPRO/TMF by 100/200 bps.

Not quite as exciting as you thought...
Hydromod
Posts: 434
Joined: Tue Mar 26, 2019 10:21 pm

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

Post by Hydromod »

MotoTrojan wrote: Tue Aug 27, 2019 4:48 pm
Hydromod wrote: Tue Aug 27, 2019 4:23 pm

Just to say that quarterly is probably about equivalent to monthly, rather than significantly better.
How about with regards to drawdown?

Annual rebalancing also seems to have merit and really is only adversely impacted in Portfolio Visualizer starting dates at-least by Black Monday.
I'd have to check on drawdown. But starting date makes a big difference for events such as Black Monday; the set of tracks for the portfolio total are approximately parallel almost all of the time but scatter at big events. It's not necessarily the case that a big event has the same effect for all starting dates.
wise_magus
Posts: 15
Joined: Sun Sep 17, 2017 1:20 pm

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

Post by wise_magus »

Hydromod wrote: Tue Aug 27, 2019 5:39 pm No, I was referring to the 50/50 UPRO/TMF (actually 40/60, but anywhere from 30/70 to 60/40 should have been similar). The daily rebalance for UPRO/TMF would have been superior to the monthly/quarterly UPRO/TMF by 100/200 bps.

Not quite as exciting as you thought...
But isn't this equivalent to a 3x leveraged 50/50 SPY/TLT portfolio that is rebalanced daily and with leverage reset daily?

In terms of the overall portfolio, the asset allocation is just
150% SPY
150% TLT
-200% CASH
in both scenarios, and it is reset to exactly match these percentages daily.
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

wise_magus wrote: Tue Aug 27, 2019 6:04 pm
Hydromod wrote: Tue Aug 27, 2019 5:39 pm No, I was referring to the 50/50 UPRO/TMF (actually 40/60, but anywhere from 30/70 to 60/40 should have been similar). The daily rebalance for UPRO/TMF would have been superior to the monthly/quarterly UPRO/TMF by 100/200 bps.

Not quite as exciting as you thought...
But isn't this equivalent to a 3x leveraged 50/50 SPY/TLT portfolio that is rebalanced daily and with leverage reset daily?

In terms of the overall portfolio, the asset allocation is just
150% SPY
150% TLT
-200% CASH
in both scenarios, and it is reset to exactly match these percentages daily.
Yes, this is what Hydro said would increase CAGR.
wise_magus
Posts: 15
Joined: Sun Sep 17, 2017 1:20 pm

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

Post by wise_magus »

MotoTrojan wrote: Tue Aug 27, 2019 5:35 pm
wise_magus wrote: Tue Aug 27, 2019 5:34 pm
Hydromod wrote: Tue Aug 27, 2019 4:23 pm The calculations I've done with different rebalance periods show that daily rebalancing mitigates some of the volatility drag, which adds a percentage point or two to the CAGR. This bonus drops off rapidly as the inverse of the number of days between rebalances. I've thought that an ETF would be the best way to implement this daily rebalance.
Just to be clear: You're saying that if there were an ETF that did a daily rebalance and daily leverage reset of a 50/50 SPY/TLT portfolio, your calculations show that be historically superior to a (monthly/quarterly rebalanced) 50/50 UPRO/TMF portfolio by something like 100-200 bps? (Or a similar statement with some other choice of ratio instead of 50/50.)

If that's the case, I wonder why such a product doesn't exist. It seems like this "excellent adventure" has become quite popular and everyone would love to get an extra 100-200 bps of return.
Pimco StocksPLUS funds do exactly this with 100% S&P500 exposure plus various bond options.
Thanks! You're right, PSLDX does this with 2x leverage, but what people might be uncomfortable with is the active management on the bond side (and the difficulty of actually purchasing PSLDX at most brokerages). If this was done using indexes for stocks and bonds, the expense ratio would also be much lower. (E.g., NTSX which does something like this with 1.5x leverage only has a 0.2% MER.)
Lee_WSP
Posts: 3995
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

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

Post by Lee_WSP »

wise_magus wrote: Tue Aug 27, 2019 6:07 pm
MotoTrojan wrote: Tue Aug 27, 2019 5:35 pm
wise_magus wrote: Tue Aug 27, 2019 5:34 pm
Hydromod wrote: Tue Aug 27, 2019 4:23 pm The calculations I've done with different rebalance periods show that daily rebalancing mitigates some of the volatility drag, which adds a percentage point or two to the CAGR. This bonus drops off rapidly as the inverse of the number of days between rebalances. I've thought that an ETF would be the best way to implement this daily rebalance.
Just to be clear: You're saying that if there were an ETF that did a daily rebalance and daily leverage reset of a 50/50 SPY/TLT portfolio, your calculations show that be historically superior to a (monthly/quarterly rebalanced) 50/50 UPRO/TMF portfolio by something like 100-200 bps? (Or a similar statement with some other choice of ratio instead of 50/50.)

If that's the case, I wonder why such a product doesn't exist. It seems like this "excellent adventure" has become quite popular and everyone would love to get an extra 100-200 bps of return.
Pimco StocksPLUS funds do exactly this with 100% S&P500 exposure plus various bond options.
Thanks! You're right, PSLDX does this with 2x leverage, but what people might be uncomfortable with is the active management on the bond side (and the difficulty of actually purchasing PSLDX at most brokerages). If this was done using indexes for stocks and bonds, the expense ratio would also be much lower. (E.g., NTSX which does something like this with 1.5x leverage only has a 0.2% MER.)
It also has a 25k buy in if you go with Vanguard; 1 Mil with Fidelity. I guess it also spits out a ton of dividends.

[deleted wrong info]
Last edited by Lee_WSP on Tue Aug 27, 2019 6:41 pm, edited 1 time in total.
HawkeyePierce
Posts: 1800
Joined: Tue Mar 05, 2019 10:29 pm
Location: Colorado

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

Post by HawkeyePierce »

Lee_WSP wrote: Tue Aug 27, 2019 6:17 pm
wise_magus wrote: Tue Aug 27, 2019 6:07 pm
MotoTrojan wrote: Tue Aug 27, 2019 5:35 pm
wise_magus wrote: Tue Aug 27, 2019 5:34 pm
Hydromod wrote: Tue Aug 27, 2019 4:23 pm The calculations I've done with different rebalance periods show that daily rebalancing mitigates some of the volatility drag, which adds a percentage point or two to the CAGR. This bonus drops off rapidly as the inverse of the number of days between rebalances. I've thought that an ETF would be the best way to implement this daily rebalance.
Just to be clear: You're saying that if there were an ETF that did a daily rebalance and daily leverage reset of a 50/50 SPY/TLT portfolio, your calculations show that be historically superior to a (monthly/quarterly rebalanced) 50/50 UPRO/TMF portfolio by something like 100-200 bps? (Or a similar statement with some other choice of ratio instead of 50/50.)

If that's the case, I wonder why such a product doesn't exist. It seems like this "excellent adventure" has become quite popular and everyone would love to get an extra 100-200 bps of return.
Pimco StocksPLUS funds do exactly this with 100% S&P500 exposure plus various bond options.
Thanks! You're right, PSLDX does this with 2x leverage, but what people might be uncomfortable with is the active management on the bond side (and the difficulty of actually purchasing PSLDX at most brokerages). If this was done using indexes for stocks and bonds, the expense ratio would also be much lower. (E.g., NTSX which does something like this with 1.5x leverage only has a 0.2% MER.)
It also has a 25k buy in if you go with Vanguard; 1 Mil with Fidelity.

It has also done worse than total stock funds.

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

That link compares the PIMCO fund with a total intl fund and shows PSLDX with a CAGR of 19.56% vs. VTIAX at 3.6%. Is that the link you meant to share?
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

Lee_WSP wrote: Tue Aug 27, 2019 6:17 pm

It has also done worse than total stock funds.

https://www.portfoliovisualizer.com/bac ... 0&total3=0
Huh? It slaughtered the Total International fund you showed there, but it's benchmark is and should be the S&P500... which it also slaughtered since inception...

Almost 15% CAGR vs. only 8.5% for the S&P500. Sharpe up from 0.58 to 0.84... Essentially the same drawdown. Fund is pretty amazing, but I am happier with 43/57 UPRO. Bit more equity leverage (and thus risk/return) and I know the bonds are all treasuries and not active/corporate.

https://www.portfoliovisualizer.com/bac ... 0&total3=0
columbia
Posts: 3023
Joined: Tue Aug 27, 2013 5:30 am

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

Post by columbia »

If you use the investor fund, the difference is very interesting, as it starts in 2008:

https://www.portfoliovisualizer.com/bac ... 0&total3=0
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

columbia wrote: Tue Aug 27, 2019 6:25 pm If you use the investor fund, the difference is very interesting, as it starts in 2008:

https://www.portfoliovisualizer.com/bac ... 0&total3=0
You should be evaluating this against the S&P500 or at a minimum Total US, which did much better but it still nearly doubled the CAGR.
BogleBobby
Posts: 38
Joined: Tue Jul 30, 2019 9:56 pm

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

Post by BogleBobby »

MotoTrojan wrote: Tue Aug 27, 2019 1:23 pmInteresting choice. I haven't played in Simba but this did pretty poorly from 1997-present on a quick PV spot-check. I also would ditch the TLT and just hold EDV, perhaps with more unleveraged equities and less UPRO to maintain the same overall exposure. This would reduce volatility drag and expenses and boost overall efficiency. You could also hold more unleveraged International and less (or no) unleveraged S&P500 if you want to reduce your dependence there.

Siamond showed that DZK had a ~3% drag beyond it's ideal index, so I would use caution for a buy & hold approach with that fund. UPRO does WAY better at tracking it's index.
Thanks! All good suggestions for me to consider.

I'm put off by EDV a bit because it's the only unleveraged bond fund that appears to have ever had a negative nominal return in Simba's workbook.

Image

Here's a similar chart the leveraged bond funds from 1955-2018:

Image

For backtesting purposes, EDV doesn't perform that great because of how bad it did in the 1970's (when stocks also did poorly). But maybe I should just ignore that era because interest rates were managed differently back then.
Lee_WSP
Posts: 3995
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

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

Post by Lee_WSP »

MotoTrojan wrote: Tue Aug 27, 2019 6:24 pm
Lee_WSP wrote: Tue Aug 27, 2019 6:17 pm

It has also done worse than total stock funds.

https://www.portfoliovisualizer.com/bac ... 0&total3=0
Huh? It slaughtered the Total International fund you showed there, but it's benchmark is and should be the S&P500... which it also slaughtered since inception...

Almost 15% CAGR vs. only 8.5% for the S&P500. Sharpe up from 0.58 to 0.84... Essentially the same drawdown. Fund is pretty amazing, but I am happier with 43/57 UPRO. Bit more equity leverage (and thus risk/return) and I know the bonds are all treasuries and not active/corporate.

https://www.portfoliovisualizer.com/bac ... 0&total3=0
Nope, that's not the link.

Okay, new question. Why is google showing the fund underperforming VTI?

https://www.google.com/search?rlz=1C1CH ... ab=COMPARE

Does it spit out a lot of dividends?
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

BogleBobby wrote: Tue Aug 27, 2019 6:33 pm
MotoTrojan wrote: Tue Aug 27, 2019 1:23 pmInteresting choice. I haven't played in Simba but this did pretty poorly from 1997-present on a quick PV spot-check. I also would ditch the TLT and just hold EDV, perhaps with more unleveraged equities and less UPRO to maintain the same overall exposure. This would reduce volatility drag and expenses and boost overall efficiency. You could also hold more unleveraged International and less (or no) unleveraged S&P500 if you want to reduce your dependence there.

Siamond showed that DZK had a ~3% drag beyond it's ideal index, so I would use caution for a buy & hold approach with that fund. UPRO does WAY better at tracking it's index.
Thanks! All good suggestions for me to consider.

I'm put off by EDV a bit because it's the only unleveraged bond fund that appears to have ever had a negative nominal return in Simba's workbook.

Image

Here's a similar chart the leveraged bond funds from 1955-2018:

Image

For backtesting purposes, EDV doesn't perform that great because of how bad it did in the 1970's (when stocks also did poorly). But maybe I should just ignore that era because interest rates were managed differently back then.
I noticed this as well but your plot shows it much more clearly... I can't quite understand why EDV did so much worse when it should only be around 1.5x the duration of 20 year treasuries, yet it had a much steeper drawdown. When looking at 43/57 UPRO/EDV vs. UPRO/LTT you get almost identical CAGR from 1955-2018 with LTT having a decent bump up on Sharpe/Sortino. Something to consider indeed. Would love to hear more about why EDV drewdown so much harder in the early years.
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

Lee_WSP wrote: Tue Aug 27, 2019 6:38 pm
MotoTrojan wrote: Tue Aug 27, 2019 6:24 pm
Lee_WSP wrote: Tue Aug 27, 2019 6:17 pm

It has also done worse than total stock funds.

https://www.portfoliovisualizer.com/bac ... 0&total3=0
Huh? It slaughtered the Total International fund you showed there, but it's benchmark is and should be the S&P500... which it also slaughtered since inception...

Almost 15% CAGR vs. only 8.5% for the S&P500. Sharpe up from 0.58 to 0.84... Essentially the same drawdown. Fund is pretty amazing, but I am happier with 43/57 UPRO. Bit more equity leverage (and thus risk/return) and I know the bonds are all treasuries and not active/corporate.

https://www.portfoliovisualizer.com/bac ... 0&total3=0
Nope, that's not the link.

Okay, new question. Why is google showing the fund underperforming VTI?

https://www.google.com/search?rlz=1C1CH ... ab=COMPARE

Does it spit out a lot of dividends?
Yes. Should never be held in taxable.
wise_magus
Posts: 15
Joined: Sun Sep 17, 2017 1:20 pm

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

Post by wise_magus »

MotoTrojan wrote: Tue Aug 27, 2019 6:06 pm
wise_magus wrote: Tue Aug 27, 2019 6:04 pm
Hydromod wrote: Tue Aug 27, 2019 5:39 pm No, I was referring to the 50/50 UPRO/TMF (actually 40/60, but anywhere from 30/70 to 60/40 should have been similar). The daily rebalance for UPRO/TMF would have been superior to the monthly/quarterly UPRO/TMF by 100/200 bps.

Not quite as exciting as you thought...
But isn't this equivalent to a 3x leveraged 50/50 SPY/TLT portfolio that is rebalanced daily and with leverage reset daily?

In terms of the overall portfolio, the asset allocation is just
150% SPY
150% TLT
-200% CASH
in both scenarios, and it is reset to exactly match these percentages daily.
Yes, this is what Hydro said would increase CAGR.
Ah, right. Sorry, I misunderstood what Hydromod wrote. All good. So daily rebalancing is great for returns.
Lee_WSP
Posts: 3995
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

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

Post by Lee_WSP »

Hydromod wrote: Tue Aug 27, 2019 5:39 pm
wise_magus wrote: Tue Aug 27, 2019 5:34 pm
Hydromod wrote: Tue Aug 27, 2019 4:23 pm The calculations I've done with different rebalance periods show that daily rebalancing mitigates some of the volatility drag, which adds a percentage point or two to the CAGR. This bonus drops off rapidly as the inverse of the number of days between rebalances. I've thought that an ETF would be the best way to implement this daily rebalance.
Just to be clear: You're saying that if there were an ETF that did a daily rebalance and daily leverage reset of a 50/50 SPY/TLT portfolio, your calculations show that be historically superior to a (monthly/quarterly rebalanced) 50/50 UPRO/TMF portfolio by something like 100-200 bps? (Or a similar statement with some other choice of ratio instead of 50/50.)

If that's the case, I wonder why such a product doesn't exist. It seems like this "excellent adventure" has become quite popular and everyone would love to get an extra 100-200 bps of return.
No, I was referring to the 50/50 UPRO/TMF (actually 40/60, but anywhere from 30/70 to 60/40 should have been similar). The daily rebalance for UPRO/TMF would have been superior to the monthly/quarterly UPRO/TMF by 100/200 bps.

Not quite as exciting as you thought...
What about compared to a 4/5% rebalance band?
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

Lee_WSP wrote: Tue Aug 27, 2019 7:15 pm
Hydromod wrote: Tue Aug 27, 2019 5:39 pm
wise_magus wrote: Tue Aug 27, 2019 5:34 pm
Hydromod wrote: Tue Aug 27, 2019 4:23 pm The calculations I've done with different rebalance periods show that daily rebalancing mitigates some of the volatility drag, which adds a percentage point or two to the CAGR. This bonus drops off rapidly as the inverse of the number of days between rebalances. I've thought that an ETF would be the best way to implement this daily rebalance.
Just to be clear: You're saying that if there were an ETF that did a daily rebalance and daily leverage reset of a 50/50 SPY/TLT portfolio, your calculations show that be historically superior to a (monthly/quarterly rebalanced) 50/50 UPRO/TMF portfolio by something like 100-200 bps? (Or a similar statement with some other choice of ratio instead of 50/50.)

If that's the case, I wonder why such a product doesn't exist. It seems like this "excellent adventure" has become quite popular and everyone would love to get an extra 100-200 bps of return.
No, I was referring to the 50/50 UPRO/TMF (actually 40/60, but anywhere from 30/70 to 60/40 should have been similar). The daily rebalance for UPRO/TMF would have been superior to the monthly/quarterly UPRO/TMF by 100/200 bps.

Not quite as exciting as you thought...
What about compared to a 4/5% rebalance band?
Easy to check, but I’d wager you’ll drawdown fairly hard.
Hydromod
Posts: 434
Joined: Tue Mar 26, 2019 10:21 pm

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

Post by Hydromod »

MotoTrojan wrote: Tue Aug 27, 2019 7:26 pm
Lee_WSP wrote: Tue Aug 27, 2019 7:15 pm
Hydromod wrote: Tue Aug 27, 2019 5:39 pm
wise_magus wrote: Tue Aug 27, 2019 5:34 pm
Hydromod wrote: Tue Aug 27, 2019 4:23 pm The calculations I've done with different rebalance periods show that daily rebalancing mitigates some of the volatility drag, which adds a percentage point or two to the CAGR. This bonus drops off rapidly as the inverse of the number of days between rebalances. I've thought that an ETF would be the best way to implement this daily rebalance.
Just to be clear: You're saying that if there were an ETF that did a daily rebalance and daily leverage reset of a 50/50 SPY/TLT portfolio, your calculations show that be historically superior to a (monthly/quarterly rebalanced) 50/50 UPRO/TMF portfolio by something like 100-200 bps? (Or a similar statement with some other choice of ratio instead of 50/50.)

If that's the case, I wonder why such a product doesn't exist. It seems like this "excellent adventure" has become quite popular and everyone would love to get an extra 100-200 bps of return.
No, I was referring to the 50/50 UPRO/TMF (actually 40/60, but anywhere from 30/70 to 60/40 should have been similar). The daily rebalance for UPRO/TMF would have been superior to the monthly/quarterly UPRO/TMF by 100/200 bps.

Not quite as exciting as you thought...
What about compared to a 4/5% rebalance band?
Easy to check, but I’d wager you’ll drawdown fairly hard.
A 5% band triggers quite frequently. I have played with this up to 25%, including minimum days between rebalancing. I don't see a very strong difference up to 20% bands and minimum of say 20 days between rebalancing. I'm putting together a set of plots to demonstrate.

The start date has a much larger influence in my opinion.

The returns from 30/70 through 60/40 would have been quite similar, with increased drawdowns, which may be why rebalancing sensitivity wasn't all that large.
caklim00
Posts: 2335
Joined: Mon May 26, 2008 10:09 am

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

Post by caklim00 »

MotoTrojan wrote: Tue Aug 27, 2019 6:39 pm
BogleBobby wrote: Tue Aug 27, 2019 6:33 pm
MotoTrojan wrote: Tue Aug 27, 2019 1:23 pmInteresting choice. I haven't played in Simba but this did pretty poorly from 1997-present on a quick PV spot-check. I also would ditch the TLT and just hold EDV, perhaps with more unleveraged equities and less UPRO to maintain the same overall exposure. This would reduce volatility drag and expenses and boost overall efficiency. You could also hold more unleveraged International and less (or no) unleveraged S&P500 if you want to reduce your dependence there.

Siamond showed that DZK had a ~3% drag beyond it's ideal index, so I would use caution for a buy & hold approach with that fund. UPRO does WAY better at tracking it's index.
Thanks! All good suggestions for me to consider.

I'm put off by EDV a bit because it's the only unleveraged bond fund that appears to have ever had a negative nominal return in Simba's workbook.

Image

Here's a similar chart the leveraged bond funds from 1955-2018:

Image

For backtesting purposes, EDV doesn't perform that great because of how bad it did in the 1970's (when stocks also did poorly). But maybe I should just ignore that era because interest rates were managed differently back then.
I noticed this as well but your plot shows it much more clearly... I can't quite understand why EDV did so much worse when it should only be around 1.5x the duration of 20 year treasuries, yet it had a much steeper drawdown. When looking at 43/57 UPRO/EDV vs. UPRO/LTT you get almost identical CAGR from 1955-2018 with LTT having a decent bump up on Sharpe/Sortino. Something to consider indeed. Would love to hear more about why EDV drewdown so much harder in the early years.
Definitely very concerning... I'm by no means an expert in this stuff but does it have to do with where it fell in duration length, or is there sotmething fundamentally flawed with what EDV holds. TMF would be the same as TLT just leveraged.

Is there something better to compare EDV to then LTT?
Hydromod
Posts: 434
Joined: Tue Mar 26, 2019 10:21 pm

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

Post by Hydromod »

The following figure illustrates my point about the bands.

The figure considers different rebalance strategies for the 40/60 UPROSIM/TMFSIM series, compared to quarterly rebalancing (Hedgefundie's OP strategy). This case starts with the first day of the series, and looks at moving 10-year periods starting on day 1, day 2, day 3, etc.

The gray lines are the nominal scheme. Magenta through blue are the band schemes. The orange and yellow lines are UPRO and TMF.

The time series plots are, from top to bottom
  • Tell-tale plot relative to quarterly rebalancing
  • Moving 10-year CAGR
  • Target weight for UPRO
  • Drawdown
The cumulative distribution function (CDF) plots are, from top to bottom
  • Trading days between rebalance
  • Moving 10-year CAGR
  • Moving difference between CAGR for different strategies and the quarterly strategy
  • Moving 10-year ulcer index
The boxes in the CDF plots are the average of all the points in the curve. The point where a curve crosses the 0.5 line is the median of all the points (the little lines crossing the bottom axis represent the median).

If I'm doing this right, it looks like there is not much difference between the band schemes. The CDF plot with days to rebalance suggests that with a 0.05 band, half of the rebalances would have occurred within 10 to 21 days of the previous rebalance, there was at least one interval of a year without rebalancing, and there would have been 1.5 times as many rebalances as the quarterly scheme.

I think that the nominal quarterly curve is a little misleading. If the analysis were started 30 days later, the average CAGR would have been 200 bps smaller for the nominal curve but not much affected for the band schemes. So I would be cautious reading too much into the returns one might expect.

Those doing adaptive schemes will find that band-based rebalancing will be much more frequent.

Image

Food for thought.
inatangle
Posts: 6
Joined: Mon Aug 05, 2019 5:14 pm

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

Post by inatangle »

I've done a fair bit of fooling around with different trading strategies in the few years that I've been trading. Whilst I've managed to extract some great theoretical returns from timed strategies, I'm wary of them because they presume that future events will play out the same as they have in the past.

Whilst the banded rebalancing strategy I'm implementing returns about 5% less than quarterly rebalancing, I think the adaptive approach reduces the risk of overfitting which the quarterly strategy may be inadvertently subject to. My next job is to brush up my coding skills so I can rebalance my portfolio whilst I sleep/work.

Huge thanks to OP and all of the other great contributors on this thread. You're all legends!
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

caklim00 wrote: Tue Aug 27, 2019 10:02 pm
Definitely very concerning... I'm by no means an expert in this stuff but does it have to do with where it fell in duration length, or is there sotmething fundamentally flawed with what EDV holds. TMF would be the same as TLT just leveraged.

Is there something better to compare EDV to then LTT?
It is not fair to simply say "TMF would be the same as TLT just leveraged" IMHO; the impact of "just leveraging" is pretty substantial, especially when rebalancing daily like TMF. It is not unlikely at all that yields will remain mostly unchanged and TMF will underperform both TLT and EDV due to volatility decay. The order of LTT<->EDV<->TMF in terms of total return only really holds true during rapidly changing rates, regardless of direction.

There is a pretty sharp rebound so it may just be that the yield curve between 20 and 30 years got MUCH steeper, quickly, due to the rapid rise in rates, causing the dip in rolling return, and then 20 year bonds caught up as yields peaked.
assetalloc
Posts: 74
Joined: Tue Nov 13, 2018 8:45 pm

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

Post by assetalloc »

Can you help me understand why you don't think increased volatility will generate more volatility decay on a daily-rebalanced holding?
My understanding is that the goal of these 3X ETFs to generate 3X daily movement and not necessarily long term movements. 3X leverage may do wonders during bullish periods. Nobody can know where S&P will be in next 15 years.

regarding your comment of 55 year historical data, that's past. Performance chasers should know that EM SC is best asset class in last 50+ years. So I brought EEM/EDC comparison to show - will you be able to stomach 15 years of underperformance if that happens to US/TLT 3X balanced portfolio? Just because EEM has rough 15 period, you are calling it volatility effect and not accepting 3X leverage + high costs of funds are useless.

If you take 100% UUPRO like portfolio, no matter what period you start, in 2008 it would have wiped you almost out completely (or left to ashes). What if you needed money in 2008 or 2009?

Proponents of "portfolio theory" based on TMF to reduce the blow are forgetting, that bonds and us markets could fall together in inflationary, rising rate periods.

The reason we buy uncorrelated assets is to reduce the volatility. So that we don't run out of money when we need it the most.
Lee_WSP
Posts: 3995
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

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

Post by Lee_WSP »

assetalloc wrote: Wed Aug 28, 2019 12:34 am
Can you help me understand why you don't think increased volatility will generate more volatility decay on a daily-rebalanced holding?
My understanding is that the goal of these 3X ETFs to generate 3X daily movement and not necessarily long term movements. 3X leverage may do wonders during bullish periods. Nobody can know where S&P will be in next 15 years.


If you take 100% UUPRO like portfolio, no matter what period you start, in 2008 it would have wiped you almost out completely (or left to ashes). What if you needed money in 2008 or 2009?

Proponents of "portfolio theory" based on TMF to reduce the blow are forgetting, that bonds and us markets could fall together in inflationary, rising rate periods.

The reason we buy uncorrelated assets is to reduce the volatility. So that we don't run out of money when we need it the most.
It would take a whole month of +/- 2% daily (2% up one day, 2% down the next) to lose 10% to volatility decay. After a whole year's worth, you still aren't quite down 50%. AFAIK, a year's worth of dailyl +/- with zero net movement every other day is pretty much a statistical impossibility.

No one should invest with money they need in the next ten years. And no one should be utilizing this strategy with money they may need in the next 20 years. Investing is gambling with a high probability of success. We're just putting more into the pot per round with money we don't care about. Those of us with common sense anyway.
assetalloc
Posts: 74
Joined: Tue Nov 13, 2018 8:45 pm

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

Post by assetalloc »

Lee_WSP wrote: Wed Aug 28, 2019 12:48 am

It would take a whole month of +/- 2% daily (2% up one day, 2% down the next) to lose 10% to volatility decay. After a whole year's worth, you still aren't quite down 50%. AFAIK, a year's worth of dailyl +/- with zero net movement every other day is pretty much a statistical impossibility.
before you hide the effect of leverage with TMF, just look at UPRO simulation just for 2008, what happened?
volatility decay causes leveraged funds to underperform even more when you look at both up and down cycle. I don't see your point on why decay is good and causes statistically impossible returns?

the debate is not about daily movements leading to no change for the year. But case when you have -25% year followed with few more bad years, and then stalled recovery for 5 years.

Regarding your comments on "long term", In my experience, usually underperforming ETFs get abandoned by investors, and ETFs get shut down. And investors lose money for good.
HawkeyePierce
Posts: 1800
Joined: Tue Mar 05, 2019 10:29 pm
Location: Colorado

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

Post by HawkeyePierce »

assetalloc wrote: Wed Aug 28, 2019 1:21 am
Lee_WSP wrote: Wed Aug 28, 2019 12:48 am

It would take a whole month of +/- 2% daily (2% up one day, 2% down the next) to lose 10% to volatility decay. After a whole year's worth, you still aren't quite down 50%. AFAIK, a year's worth of dailyl +/- with zero net movement every other day is pretty much a statistical impossibility.
before you hide the effect of leverage with TMF, just look at UPRO simulation just for 2008, what happened?
volatility decay causes leveraged funds to underperform even more when you look at both up and down cycle. I don't see your point on why decay is good and causes statistically impossible returns?

the debate is not about daily movements leading to no change for the year. But case when you have -25% year followed with few more bad years, and then stalled recovery for 5 years.

Regarding your comments on "long term", In my experience, usually underperforming ETFs get abandoned by investors, and ETFs get shut down. And investors lose money for good.
Investors don't lose their money for good when an ETF gets shut down. They get the cash value of the securities on the final day of trading (ish).
User avatar
mikestorm
Posts: 380
Joined: Tue Feb 23, 2010 8:34 am

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

Post by mikestorm »

Also, the architects of these 3x ETFs intended for them to be viewed through the lens of a single day. Long-term trends of these funds are irrelevant to them. Additionally many day-traders short these to boot, so up or down these are making money for somebody, and have value to the architects.

Even if these funds lose 90% of their NAV from today, they are no less useful as investment tools. Such a trend would only be catastrophic to those of us who are 'misusing' these ETFs by holding long term, who I am assuming are in the minority.
User avatar
Forester
Posts: 1847
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

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

Post by Forester »

mikestorm wrote: Wed Aug 28, 2019 5:54 am Such a trend would only be catastrophic to those of us who are 'misusing' these ETFs by holding long term, who I am assuming are in the minority.
Isn't that how people here are using them. I thought the danger of holding these had been overstated?
User avatar
coingaroo
Posts: 98
Joined: Fri Apr 26, 2019 11:31 am

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

Post by coingaroo »

stipeman wrote: Sat Aug 17, 2019 10:37 am\
If you have a lot of tax advantaged space you are much better off with PSLDX. Problem solved and you can spend your time on other things. Perhaps not as exciting though.
What if you don't have enough tax advantaged space - is there another easier option?
User avatar
coingaroo
Posts: 98
Joined: Fri Apr 26, 2019 11:31 am

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

Post by coingaroo »

MotoTrojan wrote: Tue Aug 27, 2019 3:40 pm Since inception EDV has outperformed TLT. VEDTX is the mutual fund version and has been around a bit longer and using that in PV you'll see even more outperformance since inception. Since 1982, when rates started dropping, EDV would've crushed TLT, but of course TMF would've been the champion.

I never said EDV had similar returns to TMF, but it does have more duration exposure than TLT (about 1/2 of TMF, or 1.5x TLT). Thus it is a way to get some "leverage" without any of the high expenses, interest, or volatility decay of TMF. It is less duration exposure than TMF, hence why I use 43/57 UPRO/EDV. 55/45 UPRO/EDV would have less flight-to-safety protection in a downturn.

It is easier to think about daily returns. In a completely flat yield curve (from 3-month libor that TMF borrows from, to 20 and 30 year bonds) TLT, EDV, and TMF will all provide the same amount of income (TMF provides 3x TLT - 2x libor, which in this example is just 1x TLT). In one day of trading, if TLT goes up 1% due to dropping rates, EDV would go up ~1.5% and TMF 3%. If TLT goes down 1%, the others would follow.

With a steeper yield curve you'll get a similar change in daily prices, with EDV providing more income that TLT, and TMF's income being somewhere around EDV (higher or lower) depending on how libor compares to the 20 year. I think the flat yield curve example (similar to today's situation anyways) makes it a lot more clear how they all compare though.
This just gave me an idea: Is there a role for EDV combined with long term corporate bonds for duration + credit risk, in order to reach a similar risk exposure (and hence overall return exposure). Yes, there is the obvious downside that credit risk is correlated with equity risk, but imperfectly. There could still be some slight diversification benefit.

I tested it with PV, and the answer is no: VCLT is a poor choice to partially replace EDV.

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

And for assetalloc, if you are concerned about stagflation, you could consider adding real assets like gold and real estate. See: Permanent Portfolio, and Ray Dalio's All Weather allocations.
User avatar
mikestorm
Posts: 380
Joined: Tue Feb 23, 2010 8:34 am

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

Post by mikestorm »

Forester wrote: Wed Aug 28, 2019 6:10 am
mikestorm wrote: Wed Aug 28, 2019 5:54 am Such a trend would only be catastrophic to those of us who are 'misusing' these ETFs by holding long term, who I am assuming are in the minority.
Isn't that how people here are using them. I thought the danger of holding these had been overstated?
Exactly. Someone pondered what would happen if the ETF lost so much value that it closed. My point is value is not necessarily a functional requirement of these ETFs.
rascott
Posts: 2450
Joined: Wed Apr 15, 2015 10:53 am

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

Post by rascott »

assetalloc wrote: Wed Aug 28, 2019 1:21 am
Lee_WSP wrote: Wed Aug 28, 2019 12:48 am

It would take a whole month of +/- 2% daily (2% up one day, 2% down the next) to lose 10% to volatility decay. After a whole year's worth, you still aren't quite down 50%. AFAIK, a year's worth of dailyl +/- with zero net movement every other day is pretty much a statistical impossibility.
before you hide the effect of leverage with TMF, just look at UPRO simulation just for 2008, what happened?
volatility decay causes leveraged funds to underperform even more when you look at both up and down cycle. I don't see your point on why decay is good and causes statistically impossible returns?

the debate is not about daily movements leading to no change for the year. But case when you have -25% year followed with few more bad years, and then stalled recovery for 5 years.

Regarding your comments on "long term", In my experience, usually underperforming ETFs get abandoned by investors, and ETFs get shut down. And investors lose money for good.

These LETFs won't underperform...even if they went to nearly zero they are not underperforming...they are doing exactly what they are supposed to do.
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

assetalloc wrote: Wed Aug 28, 2019 1:21 am
Lee_WSP wrote: Wed Aug 28, 2019 12:48 am

It would take a whole month of +/- 2% daily (2% up one day, 2% down the next) to lose 10% to volatility decay. After a whole year's worth, you still aren't quite down 50%. AFAIK, a year's worth of dailyl +/- with zero net movement every other day is pretty much a statistical impossibility.
before you hide the effect of leverage with TMF, just look at UPRO simulation just for 2008, what happened?
volatility decay causes leveraged funds to underperform even more when you look at both up and down cycle. I don't see your point on why decay is good and causes statistically impossible returns?

the debate is not about daily movements leading to no change for the year. But case when you have -25% year followed with few more bad years, and then stalled recovery for 5 years.

Regarding your comments on "long term", In my experience, usually underperforming ETFs get abandoned by investors, and ETFs get shut down. And investors lose money for good.
UPRO did great in 2008, losing FAR less than 3x the S&P500 (you didn't owe the provider anything...).

Yes TMF and UPRO could both go down, nobody is denying that. Less likely with EDV though :sharebeer .
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

coingaroo wrote: Wed Aug 28, 2019 7:42 am

This just gave me an idea: Is there a role for EDV combined with long term corporate bonds for duration + credit risk, in order to reach a similar risk exposure (and hence overall return exposure). Yes, there is the obvious downside that credit risk is correlated with equity risk, but imperfectly. There could still be some slight diversification benefit.

I tested it with PV, and the answer is no: VCLT is a poor choice to partially replace EDV.

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

And for assetalloc, if you are concerned about stagflation, you could consider adding real assets like gold and real estate. See: Permanent Portfolio, and Ray Dalio's All Weather allocations.
VCLT was a poor choice to partially replace EDV (due to falling rates and increased duration of EDV), but I bet it would line up closely with PSLDX (33/67 UPRO/EDV does too but with more duration exposure). I personally would prefer to take my risk on the equity side.
Lee_WSP
Posts: 3995
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

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

Post by Lee_WSP »

assetalloc wrote: Wed Aug 28, 2019 1:21 am
Lee_WSP wrote: Wed Aug 28, 2019 12:48 am

It would take a whole month of +/- 2% daily (2% up one day, 2% down the next) to lose 10% to volatility decay. After a whole year's worth, you still aren't quite down 50%. AFAIK, a year's worth of dailyl +/- with zero net movement every other day is pretty much a statistical impossibility.
before you hide the effect of leverage with TMF, just look at UPRO simulation just for 2008, what happened?
volatility decay causes leveraged funds to underperform even more when you look at both up and down cycle. I don't see your point on why decay is good and causes statistically impossible returns?

the debate is not about daily movements leading to no change for the year. But case when you have -25% year followed with few more bad years, and then stalled recovery for 5 years.

Regarding your comments on "long term", In my experience, usually underperforming ETFs get abandoned by investors, and ETFs get shut down. And investors lose money for good.
Yes, they're leveraged. So what? If you are personally anti leverage, that's fine. Don't tell other people what to do though.

In 2008, a 3x margin account would have been wiped out and may have even owed money if the broker didn't liquidate quickly enough. UPRO would have just kept on humming. SSO did. So, again, I am not sure what your point is. Everyone knows leverage cuts both ways, so what are you getting at? UPRO is not wiped out, just as the S&P was not wiped out.

And if the S&P was wiped out, we'd all be really really [sic].
dspencer
Posts: 225
Joined: Wed Jul 06, 2016 11:29 am

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

Post by dspencer »

Is anyone considering:

1. Lowering the total leverage because you believe it's too high for optimal returns and/or to reduce risk?
2. Adding any assets that would protect against an inflationary environment?
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

dspencer wrote: Wed Aug 28, 2019 12:51 pm Is anyone considering:

1. Lowering the total leverage because you believe it's too high for optimal returns and/or to reduce risk?
2. Adding any assets that would protect against an inflationary environment?
I have done #1 with my core holding of 43/57 UPRO/EDV. Also specifically eliminates volatility decay on the (IMHO) more susceptible holding, due to it being more likely to oscillate without any significant growth in price.
bradpevans
Posts: 619
Joined: Sun Apr 08, 2018 1:09 pm

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

Post by bradpevans »

Time to rebalance?

On March 22 (inside my Roth account FWIW) i bought two parts UPRO and three parts TMF for a 40:60 ratio

Today its more like 2 part / 5 parts, or a 28:72 ratio

This chart starts with March 22, 2019
Image

Mid May, early July, early August (wherever the trends cross) then my account was at the original 40:60 balance, so much of this is the market behavior of the just the last few weeks

Curious what "bands" people use for this.

Visually (and perhaps in hindsite), the vertical gap between the trends may
represent what is/was optimal re-balance dates. This of course is reflected in the relative
proportions of each, but seems "more obvious" in this plot above (which includes hindsight of knowing that "the gap narrowed" after reaching a peak gap

Thanks
robertmcd
Posts: 554
Joined: Tue Aug 09, 2016 9:06 am

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

Post by robertmcd »

MotoTrojan wrote: Wed Aug 28, 2019 1:10 pm
dspencer wrote: Wed Aug 28, 2019 12:51 pm Is anyone considering:

1. Lowering the total leverage because you believe it's too high for optimal returns and/or to reduce risk?
2. Adding any assets that would protect against an inflationary environment?
I have done #1 with my core holding of 43/57 UPRO/EDV. Also specifically eliminates volatility decay on the (IMHO) more susceptible holding, due to it being more likely to oscillate without any significant growth in price.
So you are saying TMF is more susceptible to the pitfalls of daily leveraged ETF's vs. UPRO?

Also, when the yield curve is steeper, would it be better to use TMF to earn the spread along with VTI? Also interested in advice regarding the best ways to utilize this strategy in an account with large taxable space compared to tax advantaged space. I am guessing it would be using TMF in tax advantaged and then VTI in my taxable account vs. the UPRO/EDV strategy.
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

robertmcd wrote: Wed Aug 28, 2019 1:46 pm
MotoTrojan wrote: Wed Aug 28, 2019 1:10 pm
dspencer wrote: Wed Aug 28, 2019 12:51 pm Is anyone considering:

1. Lowering the total leverage because you believe it's too high for optimal returns and/or to reduce risk?
2. Adding any assets that would protect against an inflationary environment?
I have done #1 with my core holding of 43/57 UPRO/EDV. Also specifically eliminates volatility decay on the (IMHO) more susceptible holding, due to it being more likely to oscillate without any significant growth in price.
So you are saying TMF is more susceptible to the pitfalls of daily leveraged ETF's vs. UPRO?

Also, when the yield curve is steeper, would it be better to use TMF to earn the spread along with VTI? Also interested in advice regarding the best ways to utilize this strategy in an account with large taxable space compared to tax advantaged space. I am guessing it would be using TMF in tax advantaged and then VTI in my taxable account vs. the UPRO/EDV strategy.
You won't get much leverage with VTI and TMF so I would wager TMF is a poor choice there; TMF is really only advantageous if you need to offset a leveraged equity position. Not sure what it has to do with a steep yield curve, I guess because you are then getting some additional income on the spread above the borrowing rate?

I haven't proven this, but yes I am saying that TMF seems more susceptible to the pitfalls. Someone earlier in this thread did an awesome analysis of UPRO and showed that when equities are going up OR down steadily, the volatility decay actually helps you (get >3x upside and <3x downside) but when the market is flat and volatile you start to decay. I don't think it is unfair to say that yields are more likely to have prolonged periods of volatility without much long-term change, than equities. This is also apparent in UPRO and TMF since inception; I believe UPRO is at around 5x the S&P500 return while TMF is closer to 1.5x 20 year bonds (this was a while back, haven't rerun the numbers, but the point stands).

If I were trying to run this with some taxable exposure I would probably hold only EDV in taxable and then UPRO with enough EDV for rebalancing in the IRA; this would still need quite a bit of EDV in the IRA though historically.
Lee_WSP
Posts: 3995
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

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

Post by Lee_WSP »

Bond funds are much more likely to have flat periods where there is very little net change.

But LTT funds are quite spiky.
Last edited by Lee_WSP on Wed Aug 28, 2019 2:06 pm, edited 1 time in total.
Hydromod
Posts: 434
Joined: Tue Mar 26, 2019 10:21 pm

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

Post by Hydromod »

bradpevans wrote: Wed Aug 28, 2019 1:45 pm Time to rebalance?

On March 22 (inside my Roth account FWIW) i bought two parts UPRO and three parts TMF for a 40:60 ratio

Today its more like 2 part / 5 parts, or a 28:72 ratio

This chart starts with March 22, 2019
Image

Mid May, early July, early August (wherever the trends cross) then my account was at the original 40:60 balance, so much of this is the market behavior of the just the last few weeks

Curious what "bands" people use for this.

Visually (and perhaps in hindsite), the vertical gap between the trends may
represent what is/was optimal re-balance dates. This of course is reflected in the relative
proportions of each, but seems "more obvious" in this plot above (which includes hindsight of knowing that "the gap narrowed" after reaching a peak gap

Thanks
My post a few back posting.php?mode=quote&f=10&p=4722555#pr4721525 suggests that using bands between 5 and 20% historically would have performed approximately the same. Presumably the triggers are due to the same large event. So I would say whatever you feel comfortable with in this range...
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

Lee_WSP wrote: Wed Aug 28, 2019 2:04 pm Bond funds are much more likely to have flat periods where there is very little net change.

But LTT funds are quite spiky.
Spikey and returning to the same place is exactly what LETF's hate, hence my thesis.
rascott
Posts: 2450
Joined: Wed Apr 15, 2015 10:53 am

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

Post by rascott »

MotoTrojan wrote: Wed Aug 28, 2019 2:26 pm
Lee_WSP wrote: Wed Aug 28, 2019 2:04 pm Bond funds are much more likely to have flat periods where there is very little net change.

But LTT funds are quite spiky.
Spikey and returning to the same place is exactly what LETF's hate, hence my thesis.
Isn't the point (or one of them) of the quarterly rebalance to counteract that?
Lee_WSP
Posts: 3995
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

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

Post by Lee_WSP »

MotoTrojan wrote: Wed Aug 28, 2019 2:26 pm
Lee_WSP wrote: Wed Aug 28, 2019 2:04 pm Bond funds are much more likely to have flat periods where there is very little net change.

But LTT funds are quite spiky.
Spikey and returning to the same place is exactly what LETF's hate, hence my thesis.
It's been a spiky up trend. Probably explains why it only tracks at 2.5x
MotoTrojan
Posts: 10779
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan »

rascott wrote: Wed Aug 28, 2019 2:28 pm
MotoTrojan wrote: Wed Aug 28, 2019 2:26 pm
Lee_WSP wrote: Wed Aug 28, 2019 2:04 pm Bond funds are much more likely to have flat periods where there is very little net change.

But LTT funds are quite spiky.
Spikey and returning to the same place is exactly what LETF's hate, hence my thesis.
Isn't the point (or one of them) of the quarterly rebalance to counteract that?
Yes but it can't counteract it 100%.
samsdad
Posts: 758
Joined: Sat Jan 02, 2016 6:20 pm

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

Post by samsdad »

deleted
Post Reply