HEDGEFUNDIE's excellent adventure Part II: The next journey

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
RayKeynes
Posts: 54
Joined: Mon Nov 11, 2019 2:14 am

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

Post by RayKeynes » Mon Nov 11, 2019 11:15 am

Lee_WSP wrote:
Mon Nov 11, 2019 11:01 am
MotoTrojan wrote:
Mon Nov 11, 2019 10:36 am
Stef wrote:
Mon Nov 11, 2019 4:15 am
@HEDGEFUNDIE
How would you approach this with almost no assets and a high savings rate? I was thinking about 70% VT + 30% SSO for the next 25-30 years, no bonds.

But how would you rebalance such a portfolio? Or not rebalance it at all and let things play out?
Volatility decay kills these products unless they are regularly rebalanced with imperfectly correlated holdings. I would not do this.
I do not subscribe to the volatility decay boogeyman, but it is real. I think the true benefit to rebalancing is selling the bonds portion and buying stocks when stocks take a huge nosedive.
Please read the text of user "Stef" again. His question was risen regarding a strategiy 70% VT and 30% SSO and NOT the here mostly discussed 55% UPRO / 45% TMF strategy. Rebalancing does make sense as it needs to be done to keep the TMF/UPRO strategy working as these are both (-1)-correlated assets. You need to perform it in the TMF/URPO as otherwise the strategy would not be the same anymore.

However, for a simple 70% VT 30% SSO approach, your'e explanation doesnt make any sense to me.

Lee_WSP
Posts: 1209
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

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

Post by Lee_WSP » Mon Nov 11, 2019 11:25 am

RayKeynes wrote:
Mon Nov 11, 2019 11:15 am
Lee_WSP wrote:
Mon Nov 11, 2019 11:01 am
MotoTrojan wrote:
Mon Nov 11, 2019 10:36 am
Stef wrote:
Mon Nov 11, 2019 4:15 am
@HEDGEFUNDIE
How would you approach this with almost no assets and a high savings rate? I was thinking about 70% VT + 30% SSO for the next 25-30 years, no bonds.

But how would you rebalance such a portfolio? Or not rebalance it at all and let things play out?
Volatility decay kills these products unless they are regularly rebalanced with imperfectly correlated holdings. I would not do this.
I do not subscribe to the volatility decay boogeyman, but it is real. I think the true benefit to rebalancing is selling the bonds portion and buying stocks when stocks take a huge nosedive.
Please read the text of user "Stef" again. His question was risen regarding a strategiy 70% VT and 30% SSO and NOT the here mostly discussed 55% UPRO / 45% TMF strategy. Rebalancing does make sense as it needs to be done to keep the TMF/UPRO strategy working as these are both (-1)-correlated assets. You need to perform it in the TMF/URPO as otherwise the strategy would not be the same anymore.

However, for a simple 70% VT 30% SSO approach, your'e explanation doesnt make any sense to me.
I think you meant to quote just MotoTrojan. I was only responding to the general comment about volatility decay. I take no stance on Stef's strategy.

User avatar
Stef
Posts: 100
Joined: Thu Oct 10, 2019 10:13 am

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

Post by Stef » Mon Nov 11, 2019 11:42 am

Volatility decay is getting less than 2/3x the performance of the SP500 in x days/weeks/months, nothing more. The SSO was "theoretically" able to recover in any crash from the last 100 years and still outperform the underlying index. Even if the SP500 goes down -2% for 2 months straight every day resulting in a massive crash and losing -70%, SSO wouldn't go to 0, there would still be almost 10% left.

If we assume that the market will go up similiar to the past (otherwise we wouldn't invest in equities at all?), a 2x leveraged ETF should outperform the index in 25-30 years in almost all scenarios yet to come.

MotoTrojan
Posts: 6936
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Mon Nov 11, 2019 12:29 pm

Stef wrote:
Mon Nov 11, 2019 10:56 am
MotoTrojan wrote:
Mon Nov 11, 2019 10:36 am
Volatility decay kills these products unless they are regularly rebalanced with imperfectly correlated holdings. I would not do this.
That's not true. You can do a backtest from 1927-2019 with 25 year periods of monthly investing. Only 12 out of 67 periods SSO would have underperformed the SP500, but in all cases only slightly. You could extend it a bit to always get an outperformance.

There is no "kill" with 2x leverage. This is not a margin.
Then why not hold 100% SSO?

User avatar
Stef
Posts: 100
Joined: Thu Oct 10, 2019 10:13 am

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

Post by Stef » Mon Nov 11, 2019 12:44 pm

MotoTrojan wrote:
Mon Nov 11, 2019 12:29 pm
Then why not hold 100% SSO?
Because:

- We don't know if USA will keep outperforming the rest of the world. Japan had once the highest market cap in the world, where are they now?
- What looks easy on paper might end up impossible to endure in reallife. What are the psychological effects of seeing your portfolio going down from 1 million to 200k? Will you really stay the course?
- What happens when you lose your job due to a big recession or even depression? You sell in the bottom.

MotoTrojan
Posts: 6936
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Mon Nov 11, 2019 12:48 pm

Stef wrote:
Mon Nov 11, 2019 12:44 pm
MotoTrojan wrote:
Mon Nov 11, 2019 12:29 pm
Then why not hold 100% SSO?
Because:

- We don't know if USA will keep outperforming the rest of the world. Japan had once the highest market cap in the world, where are they now?
- What looks easy on paper might end up impossible to endure in reallife. What are the psychological effects of seeing your portfolio going down from 1 million to 200k? Will you really stay the course?
- What happens when you lose your job due to a big recession or even depression? You sell in the bottom.
Then why VT? Would it not make more sense to hold VXUS (Total International) and then use solely SSO for your US exposure? If you feel 2x S&P500 is better than unleveraged, why hold any unleveraged US.

User avatar
Stef
Posts: 100
Joined: Thu Oct 10, 2019 10:13 am

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

Post by Stef » Mon Nov 11, 2019 12:56 pm

Because there is still a slight chance that the whole thing will fail. Who knows, maybe we'll have 10-20 years of sideways stock markets with higher volatility than expected? This might put the SSO down to a level where it would take ages to recover. The future will be different than the past. We can make the calculations and look at the last 100 years, but there is no guarantee for success.

That's why I'm having this idea with 70-75% VT and 25-30% SSO.

MotoTrojan
Posts: 6936
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Mon Nov 11, 2019 1:00 pm

Stef wrote:
Mon Nov 11, 2019 12:56 pm
Because there is still a slight chance that the whole thing will fail. Who knows, maybe we'll have 10-20 years of sideways stock markets with higher volatility than expected? This might put the SSO down to a level where it would take ages to recover. The future will be different than the past. We can make the calculations and look at the last 100 years, but there is no guarantee for success.

That's why I'm having this idea with 70-75% VT and 25-30% SSO.
Sounds good. Just pushing a bit but seems well thought out. Beyond my variant of this adventure I also hope to beat the broad market over the next 30-40 years but I’m using a small-value tilt instead of leverage.

User avatar
Stef
Posts: 100
Joined: Thu Oct 10, 2019 10:13 am

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

Post by Stef » Mon Nov 11, 2019 1:21 pm

MotoTrojan wrote:
Mon Nov 11, 2019 1:00 pm
Sounds good. Just pushing a bit but seems well thought out. Beyond my variant of this adventure I also hope to beat the broad market over the next 30-40 years but I’m using a small-value tilt instead of leverage.
It's not like I'm experienced. I just started investing. I read books, articles and watched a lot of videos. First I understood the whole passive investing thing. Then I read a lot about factor investing. Then I came across leveraged ETFs and saw the performance of UPRO and TQQQ of the last decade. Then I saw Hedgefundies thread and made the calculations with a friend.

Now I'm asking myself: can you take more risks and have a very high probability of higher returns? Without jeopardizing your whole financial independence goal? How would you setup such a portfolio and monthly savings plan? How, when and should you even rebalance?

User avatar
Forester
Posts: 564
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

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

Post by Forester » Mon Nov 11, 2019 4:04 pm

Stef wrote:
Mon Nov 11, 2019 12:56 pm
Because there is still a slight chance that the whole thing will fail. Who knows, maybe we'll have 10-20 years of sideways stock markets with higher volatility than expected? This might put the SSO down to a level where it would take ages to recover. The future will be different than the past. We can make the calculations and look at the last 100 years, but there is no guarantee for success.

That's why I'm having this idea with 70-75% VT and 25-30% SSO.
Your line of thinking is reasonable. You want to juice returns appreciably, but not bet the house on the approach, and re-balance into "normal" ETFs to manage volatility & the risk of the leveraged ETF blowing up. It's a good idea.

Lee_WSP
Posts: 1209
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

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

Post by Lee_WSP » Mon Nov 11, 2019 5:57 pm

Stef wrote:
Mon Nov 11, 2019 1:21 pm
MotoTrojan wrote:
Mon Nov 11, 2019 1:00 pm
Sounds good. Just pushing a bit but seems well thought out. Beyond my variant of this adventure I also hope to beat the broad market over the next 30-40 years but I’m using a small-value tilt instead of leverage.
It's not like I'm experienced. I just started investing. I read books, articles and watched a lot of videos. First I understood the whole passive investing thing. Then I read a lot about factor investing. Then I came across leveraged ETFs and saw the performance of UPRO and TQQQ of the last decade. Then I saw Hedgefundies thread and made the calculations with a friend.

Now I'm asking myself: can you take more risks and have a very high probability of higher returns? Without jeopardizing your whole financial independence goal? How would you setup such a portfolio and monthly savings plan? How, when and should you even rebalance?
Yes, but you need to be able to withstand a decade plus of severe underperformance and possible brink of wipeout.

tyou
Posts: 6
Joined: Fri Nov 08, 2019 9:43 am

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

Post by tyou » Tue Nov 12, 2019 6:08 am

Just a heads up on anyone trying this strategy with leverage funds in the UK - the Wisdomtree funds for 3x S&P500 and FTSE100 have charges over 3.5% p.a.!!

Instead if you can find the leverage funds from SG their charges are 0.75% which I think are better for buy and hold

User avatar
Forester
Posts: 564
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

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

Post by Forester » Tue Nov 12, 2019 8:01 am

tyou wrote:
Tue Nov 12, 2019 6:08 am
Just a heads up on anyone trying this strategy with leverage funds in the UK - the Wisdomtree funds for 3x S&P500 and FTSE100 have charges over 3.5% p.a.!!

Instead if you can find the leverage funds from SG their charges are 0.75% which I think are better for buy and hold
No it is 0.75%. All the fees are here https://www.wisdomtree.eu/en-gb/etps/equities

tyou
Posts: 6
Joined: Fri Nov 08, 2019 9:43 am

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

Post by tyou » Tue Nov 12, 2019 9:00 am

Forester wrote:
Tue Nov 12, 2019 8:01 am
tyou wrote:
Tue Nov 12, 2019 6:08 am
Just a heads up on anyone trying this strategy with leverage funds in the UK - the Wisdomtree funds for 3x S&P500 and FTSE100 have charges over 3.5% p.a.!!

Instead if you can find the leverage funds from SG their charges are 0.75% which I think are better for buy and hold
No it is 0.75%. All the fees are here https://www.wisdomtree.eu/en-gb/etps/equities
Look at this here ! Scroll down to ongoing cost

https://www.youinvest.co.uk/market-rese ... fault.aspx

MotoTrojan
Posts: 6936
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Tue Nov 12, 2019 9:51 am

tyou wrote:
Tue Nov 12, 2019 9:00 am
Forester wrote:
Tue Nov 12, 2019 8:01 am
tyou wrote:
Tue Nov 12, 2019 6:08 am
Just a heads up on anyone trying this strategy with leverage funds in the UK - the Wisdomtree funds for 3x S&P500 and FTSE100 have charges over 3.5% p.a.!!

Instead if you can find the leverage funds from SG their charges are 0.75% which I think are better for buy and hold
No it is 0.75%. All the fees are here https://www.wisdomtree.eu/en-gb/etps/equities
Look at this here ! Scroll down to ongoing cost

https://www.youinvest.co.uk/market-rese ... fault.aspx
Interest charges perhaps?

tyou
Posts: 6
Joined: Fri Nov 08, 2019 9:43 am

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

Post by tyou » Tue Nov 12, 2019 10:11 am

MotoTrojan wrote:
Tue Nov 12, 2019 9:51 am
tyou wrote:
Tue Nov 12, 2019 9:00 am
Forester wrote:
Tue Nov 12, 2019 8:01 am
tyou wrote:
Tue Nov 12, 2019 6:08 am
Just a heads up on anyone trying this strategy with leverage funds in the UK - the Wisdomtree funds for 3x S&P500 and FTSE100 have charges over 3.5% p.a.!!

Instead if you can find the leverage funds from SG their charges are 0.75% which I think are better for buy and hold
No it is 0.75%. All the fees are here https://www.wisdomtree.eu/en-gb/etps/equities
Look at this here ! Scroll down to ongoing cost

https://www.youinvest.co.uk/market-rese ... fault.aspx
Interest charges perhaps?
I don't know. Like Forester said everywhere else said 0.75%. Going to call and see what the broker says

Mister E
Posts: 10
Joined: Fri Sep 07, 2018 6:50 pm

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

Post by Mister E » Tue Nov 12, 2019 12:43 pm

Probably already asked? In a taxed account if at some future point one gets out of UPRO and TMF and trades into equivalent 1x ETFs is that a wash? If so, which ETFs?
Last edited by Mister E on Tue Nov 12, 2019 12:45 pm, edited 1 time in total.

chrisdds98
Posts: 106
Joined: Tue May 19, 2015 9:55 pm
Location: Austin, TX

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

Post by chrisdds98 » Tue Nov 12, 2019 12:44 pm

Stef wrote:
Mon Nov 11, 2019 12:56 pm
Because there is still a slight chance that the whole thing will fail. Who knows, maybe we'll have 10-20 years of sideways stock markets with higher volatility than expected? This might put the SSO down to a level where it would take ages to recover. The future will be different than the past. We can make the calculations and look at the last 100 years, but there is no guarantee for success.

That's why I'm having this idea with 70-75% VT and 25-30% SSO.
I'm doing something similar. my roth is 75% UPRO 25% VT. I plan to rebalance whenever UPRO gets to 80% or 70%. The roth makes up a small percentage of my total investments so I don't mind taking a bit of risk here.

Walkure
Posts: 138
Joined: Tue Apr 11, 2017 9:59 pm

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

Post by Walkure » Tue Nov 12, 2019 12:52 pm

Mister E wrote:
Tue Nov 12, 2019 12:43 pm
Probably already asked? In a taxed account if at some future point one gets out of UPRO and TMF and trades into equivalent 1x ETFs is that a wash? If so, which ETFs?
A 1x and 3x fund are not substantially identical regardless of what underlying index they track. Also, you only care about a wash sale in the case where the thing you are selling has losses, and I can't imagine why you would sell a 3x fund when it's down only to turn around and buy a 1x of the same index. Cutting your losses and going to cash, sure, but not to re-buy a "less low" version of it! I guess you could harvest a massive loss in the short term, but the way you describe this as "at some future point," you'd have been rebalancing little losses all the way down in your taxable already anyway.

Pandananana
Posts: 3
Joined: Tue Nov 05, 2019 11:13 am

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

Post by Pandananana » Wed Nov 13, 2019 2:23 am

tyou wrote:
Tue Nov 12, 2019 10:11 am
MotoTrojan wrote:
Tue Nov 12, 2019 9:51 am
tyou wrote:
Tue Nov 12, 2019 9:00 am
Forester wrote:
Tue Nov 12, 2019 8:01 am
tyou wrote:
Tue Nov 12, 2019 6:08 am
Just a heads up on anyone trying this strategy with leverage funds in the UK - the Wisdomtree funds for 3x S&P500 and FTSE100 have charges over 3.5% p.a.!!

Instead if you can find the leverage funds from SG their charges are 0.75% which I think are better for buy and hold
No it is 0.75%. All the fees are here https://www.wisdomtree.eu/en-gb/etps/equities
Look at this here ! Scroll down to ongoing cost

https://www.youinvest.co.uk/market-rese ... fault.aspx
Interest charges perhaps?
I don't know. Like Forester said everywhere else said 0.75%. Going to call and see what the broker says
Please keep us updated about what you find out. I am also interested in buying Wisdomtrees 3X S&P 500.

DaveG75
Posts: 15
Joined: Thu Oct 10, 2019 4:57 pm

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

Post by DaveG75 » Fri Nov 15, 2019 1:29 pm

Has anyone given any thought to REITs as a leveraged asset class for this strategy? Apologies if I missed it.

Portfolio visualizer gives a correlation for DRN to UPRO of 0.63. But the rolling 12-month correlation is all over the map from a high of >0.9 to a low of about -.11. Annualized returns for the two asset classes are close (at least from 8/1/2009, DRN inception date), so it's not a big drag like gold or cash would be if these are robust characteristics.

Assuming DRN is just 3x icf, PV has data to 3/1/2001, which shows similar characteristics (i.e., similar return as SPY with correlation about 0.62.

The other caveat is these reits are far more volatile. Annualized STD for DRN is 49%!

I'm considering a small allocation of the equity portion to DRN would be a better boost than TQQQ. I would appreciate any thoughts on this.

MotoTrojan
Posts: 6936
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Fri Nov 15, 2019 2:32 pm

DaveG75 wrote:
Fri Nov 15, 2019 1:29 pm
Has anyone given any thought to REITs as a leveraged asset class for this strategy? Apologies if I missed it.

Portfolio visualizer gives a correlation for DRN to UPRO of 0.63. But the rolling 12-month correlation is all over the map from a high of >0.9 to a low of about -.11. Annualized returns for the two asset classes are close (at least from 8/1/2009, DRN inception date), so it's not a big drag like gold or cash would be if these are robust characteristics.

Assuming DRN is just 3x icf, PV has data to 3/1/2001, which shows similar characteristics (i.e., similar return as SPY with correlation about 0.62.

The other caveat is these reits are far more volatile. Annualized STD for DRN is 49%!

I'm considering a small allocation of the equity portion to DRN would be a better boost than TQQQ. I would appreciate any thoughts on this.
Volatility is never a good thing with daily rebalancing funds due to volatility decay.

DaveG75
Posts: 15
Joined: Thu Oct 10, 2019 4:57 pm

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

Post by DaveG75 » Fri Nov 15, 2019 5:27 pm

MotoTrojan wrote:
Fri Nov 15, 2019 2:32 pm
Volatility is never a good thing with daily rebalancing funds due to volatility decay.
[/quote]

I'm not sure I can agree with you there. DRN has an annualized return of approximately the same as UPRO since 2009. Over that time it does show a crazy annualized volatility of 49%. It seems to me, that despite the huge drag this volatility imposes, DRN still performs well. Am I missing something?

MotoTrojan
Posts: 6936
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Fri Nov 15, 2019 5:34 pm

DaveG75 wrote:
Fri Nov 15, 2019 5:27 pm
MotoTrojan wrote:
Fri Nov 15, 2019 2:32 pm
Volatility is never a good thing with daily rebalancing funds due to volatility decay.
I'm not sure I can agree with you there. DRN has an annualized return of approximately the same as UPRO since 2009. Over that time it does show a crazy annualized volatility of 49%. It seems to me, that despite the huge drag this volatility imposes, DRN still performs well. Am I missing something?
[/quote]

Volatility decay really gets you during flatter markets. These products actually beat the naive 3x return during steady up AND down markets so perhaps the REIT bull has made up for the volatility.

Agree or not, it’s just simple math. Measuring it during a bull is flawed IMHO.

User avatar
305pelusa
Posts: 1020
Joined: Fri Nov 16, 2018 10:20 pm

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

Post by 305pelusa » Fri Nov 15, 2019 5:39 pm

DaveG75 wrote:
Fri Nov 15, 2019 5:27 pm
MotoTrojan wrote:
Fri Nov 15, 2019 2:32 pm
Volatility is never a good thing with daily rebalancing funds due to volatility decay.
I'm not sure I can agree with you there. DRN has an annualized return of approximately the same as UPRO since 2009. Over that time it does show a crazy annualized volatility of 49%. It seems to me, that despite the huge drag this volatility imposes, DRN still performs well. Am I missing something?
[/quote]

I'm confused. UPRO's annualized returns have absolutely crushed DRN's since inception. Most of that outperformance is recent though. In fact, from Aug 2009 to oct 2016, they have identical returns.

Now that we have a time period, we can assess what the damage of volatility decay is. Take a look at whatever DRN's unleveraged equivalent (ICF you say?) against SPY in that same time period. The amount by which DRN's unleveraged etf beats SPY is precisely the volatility decay.

DaveG75
Posts: 15
Joined: Thu Oct 10, 2019 4:57 pm

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

Post by DaveG75 » Fri Nov 15, 2019 7:23 pm

305pelusa wrote:
Fri Nov 15, 2019 5:39 pm
DaveG75 wrote:
Fri Nov 15, 2019 5:27 pm
MotoTrojan wrote:
Fri Nov 15, 2019 2:32 pm

Volatility is never a good thing with daily rebalancing funds due to volatility decay.
I'm not sure I can agree with you there. DRN has an annualized return of approximately the same as UPRO since 2009. Over that time it does show a crazy annualized volatility of 49%. It seems to me, that despite the huge drag this volatility imposes, DRN still performs well. Am I missing something?
I'm confused. UPRO's annualized returns have absolutely crushed DRN's since inception. Most of that outperformance is recent though. In fact, from Aug 2009 to oct 2016, they have identical returns.

Now that we have a time period, we can assess what the damage of volatility decay is. Take a look at whatever DRN's unleveraged equivalent (ICF you say?) against SPY in that same time period. The amount by which DRN's unleveraged etf beats SPY is precisely the volatility decay.
ICF is just an etf that tracks a REIT index with an old inception date.
ICF tracks Cohen & Steers Realty Majors Index
DRN tracks The MSCI US IMI Real Estate 25/50 Index, launched on Sep 01, 2016.

FREL is an ETF tracks the relevant index (MSCI US IMI 25/50) , but the history is so short I don't think it's useful.

As far as crushed, maybe I'm looking at the wrong data.
From 8/01/2009 to 10/31/2016, PV reports:
annualized return STDV
DRN 36.38% 53.2% :shock:
UPRO 33.17% 39.38%

From 8/01/2009 to today, PV reports:
annualized return STDV
DRN 30.54% 49.2%
UPRO 34.15% 38.44%

To be sure, approximately a 3.6% difference in annual return is pretty high, but I think you can get some benefit from non-correlation/re-balancing? :confused

For comparison, from 8/01/2009 to 10/31/2016, PV reports
annualized return STDV
ICF 10.62% 23.23%
SPY 5.51% 14.61%

Assuming that ICF *ROUGHLY* corresponds to a 1x (unlevered) DRN, then yes I can see a lot of decay. Again though, there are all sorts of difficulties with this analysis (different indices, etc).

I get what you are saying about measuring during a bull. however 8/09 to 10/16 had a few admittedly minor corrections, no?

User avatar
305pelusa
Posts: 1020
Joined: Fri Nov 16, 2018 10:20 pm

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

Post by 305pelusa » Fri Nov 15, 2019 8:02 pm

DaveG75 wrote:
Fri Nov 15, 2019 7:23 pm

As far as crushed, maybe I'm looking at the wrong data.
From 8/01/2009 to 10/31/2016, PV reports:
annualized return STDV
DRN 36.38% 53.2% :shock:
UPRO 33.17% 39.38%
Sorry, I meant to Nov 2016. It's amazing what a difference that makes in these leveraged ETFs.
DaveG75 wrote:
Fri Nov 15, 2019 7:23 pm
From 8/01/2009 to today, PV reports:
annualized return STDV
DRN 30.54% 49.2%
UPRO 34.15% 38.44%
I had it on yearly returns so it cutoff 2009 for me. Yes since inception, they're close, I agree. So you could test SPY vs ICF since Aug 2009 to today and that might be somewhat representative as well.
That said, 4% compounded for a decade is serious cash. I can see you agree.
DaveG75 wrote:
Fri Nov 15, 2019 7:23 pm

For comparison, from 8/01/2009 to 10/31/2016, PV reports
annualized return STDV
ICF 10.62% 23.23%
SPY 5.51% 14.61%

Assuming that ICF *ROUGHLY* corresponds to a 1x (unlevered) DRN, then yes I can see a lot of decay. Again though, there are all sorts of difficulties with this analysis (different indices, etc).
If you want to know how much volatility decay affects a x3 LETFs, then you need to find what the unleveraged index is. Comparing it with some other x3 ETF makes little sense; it's apples and oranges.
DaveG75 wrote:
Fri Nov 15, 2019 7:23 pm

I get what you are saying about measuring during a bull. however 8/09 to 10/16 had a few admittedly minor corrections, no?
I'm not Moto Trojan. Personally, I'm unconvinced that volatility decay is a factor of the underlying index returns. I have thus far never seen a proof of it in this thread. A lot of subjective comments like "in trending markets, there's little decay" or that it "really gets you in flat markets" as though it's effect is more corrosive then. Not sure what "simple math" he refers to but I'm not sure it's a trivial proof necessarily.

I completely agree that a REIT bull market would completely "make up" for the decay, making UPRO and DRN look similar in returns. Which is why this is apples to oranges. So account for it properly by looking at the underlying, unleveraged security and see how much more in needs to overperform for DRN to keep pace with UPRO.

You could probably mathematically quantify volatility decay. Not sure if someone has done so already in the thread.

DaveG75
Posts: 15
Joined: Thu Oct 10, 2019 4:57 pm

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

Post by DaveG75 » Fri Nov 15, 2019 9:49 pm

305pelusa wrote:
Fri Nov 15, 2019 8:02 pm
DaveG75 wrote:
Fri Nov 15, 2019 7:23 pm

As far as crushed, maybe I'm looking at the wrong data.
From 8/01/2009 to 10/31/2016, PV reports:
annualized return STDV
DRN 36.38% 53.2% :shock:
UPRO 33.17% 39.38%
Sorry, I meant to Nov 2016. It's amazing what a difference that makes in these leveraged ETFs.
DaveG75 wrote:
Fri Nov 15, 2019 7:23 pm
From 8/01/2009 to today, PV reports:
annualized return STDV
DRN 30.54% 49.2%
UPRO 34.15% 38.44%
I had it on yearly returns so it cutoff 2009 for me. Yes since inception, they're close, I agree. So you could test SPY vs ICF since Aug 2009 to today and that might be somewhat representative as well.
That said, 4% compounded for a decade is serious cash. I can see you agree.
DaveG75 wrote:
Fri Nov 15, 2019 7:23 pm

For comparison, from 8/01/2009 to 10/31/2016, PV reports
annualized return STDV
ICF 10.62% 23.23%
SPY 5.51% 14.61%

Assuming that ICF *ROUGHLY* corresponds to a 1x (unlevered) DRN, then yes I can see a lot of decay. Again though, there are all sorts of difficulties with this analysis (different indices, etc).
If you want to know how much volatility decay affects a x3 LETFs, then you need to find what the unleveraged index is. Comparing it with some other x3 ETF makes little sense; it's apples and oranges.
DaveG75 wrote:
Fri Nov 15, 2019 7:23 pm

I get what you are saying about measuring during a bull. however 8/09 to 10/16 had a few admittedly minor corrections, no?
I'm not Moto Trojan. Personally, I'm unconvinced that volatility decay is a factor of the underlying index returns. I have thus far never seen a proof of it in this thread. A lot of subjective comments like "in trending markets, there's little decay" or that it "really gets you in flat markets" as though it's effect is more corrosive then. Not sure what "simple math" he refers to but I'm not sure it's a trivial proof necessarily.

I completely agree that a REIT bull market would completely "make up" for the decay, making UPRO and DRN look similar in returns. Which is why this is apples to oranges. So account for it properly by looking at the underlying, unleveraged security and see how much more in needs to overperform for DRN to keep pace with UPRO.

You could probably mathematically quantify volatility decay. Not sure if someone has done so already in the thread.
I think we are all on the same page that volatility decay is not an issue for the underlying index, rather it arises (if at all) in the daily resetting levered products. I don't think the right data exists for DRN to make the right comparison to see the decay. As I wrote in my last post, the index underlying DRN launched in 2016.

I think URE is the better exemplar and illustrates MotoTrojan's point. URE is *only* a 2x REIT, and the index for URE is the Dow Jones U.S. Real Estate Index, which is tracked by IYR. URE inception date is 1/03/2007, so we have our downturn:) URE from inception to date has a negative annualized return. Ouch! Of course, this analysis doesn't include the effect of rebalancing.

DaveG75
Posts: 15
Joined: Thu Oct 10, 2019 4:57 pm

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

Post by DaveG75 » Fri Nov 15, 2019 10:21 pm

DaveG75 wrote:
Fri Nov 15, 2019 9:49 pm
Of course, this analysis doesn't include the effect of rebalancing.
Okay, I tested monthly rebalancing using UPROSIM TMFSIM data back to 3/2007 (oldest available data for URE, which is 2x REIT). I think this is fairly definitive. No combination of all three TMF/URE/UPRO that I could find beats 50:50 TMF UPRO. Granted, I didn't check exhaustively, but the combinations I did check were all very much worse than 50:50 UPRO.

Makes me think what we really need is 3x leveraged US minvol. Any ideas on how to do that?

User avatar
305pelusa
Posts: 1020
Joined: Fri Nov 16, 2018 10:20 pm

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

Post by 305pelusa » Fri Nov 15, 2019 10:30 pm

DaveG75 wrote:
Fri Nov 15, 2019 9:49 pm
I don't think the right data exists for DRN to make the right comparison to see the decay. As I wrote in my last post, the index underlying DRN launched in 2016.
Well that would make no sense right? DRN must have been following something before 2016 don't you think?
I decided to look around.

Turns out DRN only switched to MSCI IMI RE 25/50 literally like 2 months ago. It's on the prospectus:
http://direxioninvestments.onlineprospe ... Prospectus

Since inception until Aug 2019, DRN has been following the MSCI US REIT index. That's the same index that VNQ follows (until that one switched to the 25/50 on 2018 I think). So almost all of the data (and certainly from inception to 2016) is when DRN followed an index for which there is an unleveraged ETF and plenty of raw index data from.

So Aug 2009 to Nov 2016
DRN returned 34.75%
UPRO returned 34.64%
VNQ returned 21.79%
SPY returned 17.64%

Notice two things: (1) How much the volatility decay (and high fees of course) eat away at the LETFs of both SPY and VNQ. And (2) that DRN's volatility decay is much more pronounced than UPRO's. As we expected.

Lee_WSP
Posts: 1209
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

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

Post by Lee_WSP » Fri Nov 15, 2019 11:57 pm

Volatility decay as it is commonly understood is best exemplified by SSO vs VFINX. The 2x daily rebalanced is not even 50% higher (1.3x), whereas if you had a 2x margin at the outset, you would be up 2x before paying back the principal & interest.

MotoTrojan
Posts: 6936
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Sat Nov 16, 2019 12:50 am

305pelusa wrote:
Fri Nov 15, 2019 8:02 pm
DaveG75 wrote:
Fri Nov 15, 2019 7:23 pm

As far as crushed, maybe I'm looking at the wrong data.
From 8/01/2009 to 10/31/2016, PV reports:
annualized return STDV
DRN 36.38% 53.2% :shock:
UPRO 33.17% 39.38%
Sorry, I meant to Nov 2016. It's amazing what a difference that makes in these leveraged ETFs.
DaveG75 wrote:
Fri Nov 15, 2019 7:23 pm
From 8/01/2009 to today, PV reports:
annualized return STDV
DRN 30.54% 49.2%
UPRO 34.15% 38.44%
I had it on yearly returns so it cutoff 2009 for me. Yes since inception, they're close, I agree. So you could test SPY vs ICF since Aug 2009 to today and that might be somewhat representative as well.
That said, 4% compounded for a decade is serious cash. I can see you agree.
DaveG75 wrote:
Fri Nov 15, 2019 7:23 pm

For comparison, from 8/01/2009 to 10/31/2016, PV reports
annualized return STDV
ICF 10.62% 23.23%
SPY 5.51% 14.61%

Assuming that ICF *ROUGHLY* corresponds to a 1x (unlevered) DRN, then yes I can see a lot of decay. Again though, there are all sorts of difficulties with this analysis (different indices, etc).
If you want to know how much volatility decay affects a x3 LETFs, then you need to find what the unleveraged index is. Comparing it with some other x3 ETF makes little sense; it's apples and oranges.
DaveG75 wrote:
Fri Nov 15, 2019 7:23 pm

I get what you are saying about measuring during a bull. however 8/09 to 10/16 had a few admittedly minor corrections, no?
I'm not Moto Trojan. Personally, I'm unconvinced that volatility decay is a factor of the underlying index returns. I have thus far never seen a proof of it in this thread. A lot of subjective comments like "in trending markets, there's little decay" or that it "really gets you in flat markets" as though it's effect is more corrosive then. Not sure what "simple math" he refers to but I'm not sure it's a trivial proof necessarily.

I completely agree that a REIT bull market would completely "make up" for the decay, making UPRO and DRN look similar in returns. Which is why this is apples to oranges. So account for it properly by looking at the underlying, unleveraged security and see how much more in needs to overperform for DRN to keep pace with UPRO.

You could probably mathematically quantify volatility decay. Not sure if someone has done so already in the thread.
Someone in a prior post had an awesome plot of annual returns of S&P500 and UPRO on the X and Y axis which was an amazing visual of what I am referring to. It showed a line at a slope of 3 so you could see when UPRO over or underperformed 3x the annual S&P500 return.

As shown above though the REIT fund had insane volatility drag and only kept up with UPRO because REITs outperformed. Looks like a terrible long term hold.

User avatar
Stef
Posts: 100
Joined: Thu Oct 10, 2019 10:13 am

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

Post by Stef » Sat Nov 16, 2019 8:30 am

Image

There is really no benefit in rebalancing when you are 100% in stocks with a part in leveraged ETFs. In this graphic rebalancing was applied every 30. June and 31. December of the year. If the market crashes you are going down anyway. But you lose potential gains in the bull market afterwards.

Kbg
Posts: 38
Joined: Thu Mar 23, 2017 11:33 am

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

Post by Kbg » Sat Nov 16, 2019 3:37 pm

What did you rebalance it with?

MotoTrojan
Posts: 6936
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Sat Nov 16, 2019 4:03 pm

Kbg wrote:
Sat Nov 16, 2019 3:37 pm
What did you rebalance it with?
S&P500. Perfectly correlated.... bad example.

Kbg
Posts: 38
Joined: Thu Mar 23, 2017 11:33 am

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

Post by Kbg » Sat Nov 16, 2019 7:52 pm

Ok, I was thinking it was a pretty bizarre comparison and generating tax consequences for no good reason performance wise.

User avatar
Forester
Posts: 564
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

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

Post by Forester » Sun Nov 17, 2019 6:35 am

Playing with the CoT data. This is the two main indicators from Stephen Breise's book.

Image

The Commercial CoT Index reached a buying climax (over 90%) from April to June. This is the net position based on a three year lookback; even if they were short in absolute terms, it's the relative position versus history. The CoT Movement Index subsequently plunged, seen in the +-40% six week rate of change.

People who regularly look at the CoT data wouldn't have been surprised by the late '18 sell off.

RayKeynes
Posts: 54
Joined: Mon Nov 11, 2019 2:14 am

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

Post by RayKeynes » Mon Nov 18, 2019 5:05 am

Is there an openly available dataset of either TMF or its underlying?

I would like to add simulated TMF to my calculator backtesting to 1929 to see how it performed against UPRO / SSO and SPY.

Anyone can help me?

User avatar
Forester
Posts: 564
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

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

Post by Forester » Mon Nov 18, 2019 7:02 pm

Weekly timing the S&P 500 (SPY) using the COT report. Net commercial trader position as % of 78 week range, smoothed over last 6 weeks.

07/01/1997 to 12/11/2019.

Buy when commercials bullish (COT index over 55%)
457 trades, 57.33% winning
$1,000 becomes $2,038

Buy when commercials neutral (COT index between 31.5% & 55%)
348 trades, 63.22% winning
$1,000 becomes $3,505

Buy when commercials bearish (COT index below 31.5%)
385 trades, 55.84% winning
$1,000 becomes $881

This suggests that it could be profitable to reduce equity exposure when commercial traders are bearish, otherwise stay long. There will be better ways of using the data but it's a promising start.

User avatar
Forester
Posts: 564
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

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

Post by Forester » Mon Nov 18, 2019 7:41 pm

This time it is commercials vs small traders.

07/01/1997 to 12/11/2019.

Buy when commercials bullish & small traders bearish (Commercial COT index over 30%, Small trader COT index under 50%)
327 trades (out of 1193), 59.94% winning
$1,000 becomes $2,961

Buy when commercials bearish & small traders bullish (Commercial COT index under 30%, Small trader COT index over 50%)
287 trades (out of 1193), 54.36% winning
$1,000 becomes $681

keith6014
Posts: 243
Joined: Thu Jan 02, 2014 11:58 pm

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

Post by keith6014 » Mon Nov 18, 2019 9:25 pm

How does the CoT work? I see Dec 24th 2018 as SELL but I would buy. Is it supposed to have some predictive power?

User avatar
Forester
Posts: 564
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

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

Post by Forester » Tue Nov 19, 2019 4:39 am

keith6014 wrote:
Mon Nov 18, 2019 9:25 pm
How does the CoT work? I see Dec 24th 2018 as SELL but I would buy. Is it supposed to have some predictive power?
The Sell was the -40% rate of change of a net position, it is mis-labelled. The COT data presented like that would be useful to a discretionary trader, not a mechanical system. You can see the COT index rapidly falling a few weeks ahead of the late '18 correction.

The COT data is gathered earlier in the week and released for free on the Friday (unless there's a public holiday or government shutdown). There are three categories of traders. The theory is that because commercials are (informed) hedgers, they tend to lead market turns. In the S&P 500 non-commercials would be CTA trend followers, large hedge funds. Non reportables on average would have less money to work with and be less informed, have less informational resources, than commercials.

One interesting point is that in every proposed COT system the authors have noted the system usually performs better when the data is lagged. This would make it just as useful to people trading weekly or monthly rather than daily.

keith6014
Posts: 243
Joined: Thu Jan 02, 2014 11:58 pm

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

Post by keith6014 » Tue Nov 19, 2019 7:18 am

Forester wrote:
Tue Nov 19, 2019 4:39 am
keith6014 wrote:
Mon Nov 18, 2019 9:25 pm
How does the CoT work? I see Dec 24th 2018 as SELL but I would buy. Is it supposed to have some predictive power?
The Sell was the -40% rate of change of a net position, it is mis-labelled. The COT data presented like that would be useful to a discretionary trader, not a mechanical system. You can see the COT index rapidly falling a few weeks ahead of the late '18 correction.
The late 2018 correction started in September. If COT signaled August or even until October I would be a a believer.

RayKeynes
Posts: 54
Joined: Mon Nov 11, 2019 2:14 am

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

Post by RayKeynes » Tue Nov 19, 2019 7:58 am

I've run the TMF / UPRO simulation myself in my Portfolio Calculator and came to the following results for the period 1987 - 2019:

Calculation methodology
- investments of 3'000$ on a monthly basis depending on portfolio allocation in %
- investment of initial 10'000$ depending on portfolio allocation in %
- no rebalancing applied

Image

My question is:

- How will TMF / UPRO perform in a scenario of rising interest rates and side-ways market like from 1937 - 1950?
- Do we have any proxy data to simulate TMF from 1930 - 1980? While the period 1980 to 2020 was characterized by lower interest rates over decades (hence decreasing), the period from 1940 - 1980 was characterized by higher interest rates over decades (hence increasing). Did you guys perform academical research how this portfolio would behave in a scenario like from 1940 - 1980?

Thank you for your attention.

RayKeynes
Posts: 54
Joined: Mon Nov 11, 2019 2:14 am

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

Post by RayKeynes » Tue Nov 19, 2019 11:03 am

Here is the log scale chart including the following strategies from 1987 - 2019:

- 50% SPY / 30% SSO / 20% UPRO
- 50% SPY / 25% UPRO / 25% TMF
- 50% UPRO / 50% TMF
- 100% UPRO
- 100% TMF

Approach: 3'000 monthly investment; 10'000 initial investment

Image

MotoTrojan
Posts: 6936
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Tue Nov 19, 2019 11:49 am

RayKeynes wrote:
Tue Nov 19, 2019 7:58 am
I've run the TMF / UPRO simulation myself in my Portfolio Calculator and came to the following results for the period 1987 - 2019:

Calculation methodology
- investments of 3'000$ on a monthly basis depending on portfolio allocation in %
- investment of initial 10'000$ depending on portfolio allocation in %
- no rebalancing applied

Image

My question is:

- How will TMF / UPRO perform in a scenario of rising interest rates and side-ways market like from 1937 - 1950?
- Do we have any proxy data to simulate TMF from 1930 - 1980? While the period 1980 to 2020 was characterized by lower interest rates over decades (hence decreasing), the period from 1940 - 1980 was characterized by higher interest rates over decades (hence increasing). Did you guys perform academical research how this portfolio would behave in a scenario like from 1940 - 1980?

Thank you for your attention.
The original thread has data from 1955 to present. TMF got crushed, and by present it was nearly a tie with 100% S&P500.

RayKeynes
Posts: 54
Joined: Mon Nov 11, 2019 2:14 am

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

Post by RayKeynes » Tue Nov 19, 2019 12:01 pm

Could someone send me the data for LTT and/or TMF from 1955 onwards?

I'd like to run some tests.

User avatar
Forester
Posts: 564
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

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

Post by Forester » Tue Nov 19, 2019 1:44 pm

Since 24/03/2009, trading S&P 500 end of week.

Buy & hold +373%

52-week momentum +333%

COT Index (cash if commercial -30% AND small trader +70%) +410%

RayKeynes
Posts: 54
Joined: Mon Nov 11, 2019 2:14 am

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

Post by RayKeynes » Wed Nov 20, 2019 3:11 am

MotoTrojan wrote:
Tue Nov 19, 2019 11:49 am

The original thread has data from 1955 to present. TMF got crushed, and by present it was nearly a tie with 100% S&P500.
Two questions

Where can I find this dataset from 1955 to present?
What do you mean by "TMF got crushed"? The fund got wiped out and/or went to 0?

User avatar
firebirdparts
Posts: 370
Joined: Thu Jun 13, 2019 4:21 pm

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

Post by firebirdparts » Wed Nov 20, 2019 7:14 am

They just meant TMF, had it existed, would have a bad patch about when you’d expect. The HFEAdventure portfolio does not do well in the 1970s especially.
A fool and your money are soon partners

Post Reply