HEDGEFUNDIE's excellent adventure Part II: The next journey

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
dspencer
Posts: 195
Joined: Wed Jul 06, 2016 11:29 am

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

Post by dspencer » Mon Sep 09, 2019 8:08 am

comeinvest wrote:
Mon Sep 09, 2019 2:56 am

To everybody: What's the proposed end game of Hedgefundie's strategy (or its variations) if U.S. rates follow the pattern of Japan and Europe? Would you still play this strategy if 10-year treasuries are at -1% and 30-year at -0.5% nominal, -2.5% and -2% real interest? If not, at what point would you pull the plug, or change the strategy?
I would not stay in this strategy at that point. I'm still trying to answer the last question.

User avatar
privatefarmer
Posts: 588
Joined: Mon Sep 08, 2014 2:45 pm

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

Post by privatefarmer » Mon Sep 09, 2019 8:28 am

dspencer wrote:
Mon Sep 09, 2019 8:08 am
comeinvest wrote:
Mon Sep 09, 2019 2:56 am

To everybody: What's the proposed end game of Hedgefundie's strategy (or its variations) if U.S. rates follow the pattern of Japan and Europe? Would you still play this strategy if 10-year treasuries are at -1% and 30-year at -0.5% nominal, -2.5% and -2% real interest? If not, at what point would you pull the plug, or change the strategy?
I would not stay in this strategy at that point. I'm still trying to answer the last question.
this kind of sentiment is exactly what keeps the market "efficient". the only way yields would go negative is if the MARKET drove them negative. if the collective market of investors does not feel that negative yielding long-term debt is a worthwhile investment then they won't ever go negative.

Hydromod
Posts: 198
Joined: Tue Mar 26, 2019 10:21 pm

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

Post by Hydromod » Mon Sep 09, 2019 8:37 am

dspencer wrote:
Mon Sep 09, 2019 8:08 am
comeinvest wrote:
Mon Sep 09, 2019 2:56 am

To everybody: What's the proposed end game of Hedgefundie's strategy (or its variations) if U.S. rates follow the pattern of Japan and Europe? Would you still play this strategy if 10-year treasuries are at -1% and 30-year at -0.5% nominal, -2.5% and -2% real interest? If not, at what point would you pull the plug, or change the strategy?
I would not stay in this strategy at that point. I'm still trying to answer the last question.
I'm also still trying to answer the last question. I'm mulling over strategies for (i) backing down the leverage for some period of time, and (ii) possibly altering the mix of assets. For me, the key is to use the strategy as a tool to augment my overall portfolio when conditions are likely to be neutral to favorable.

rascott
Posts: 702
Joined: Wed Apr 15, 2015 10:53 am

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

Post by rascott » Mon Sep 09, 2019 8:53 am

MinhN wrote:
Sun Sep 08, 2019 5:56 pm
I am considering using a modified version of this. With 1/3 in UPRO and 2/3 in bonds. It's like going 100% in equity with the added bonus of having extra bond income to offset the expense of running a leveraged fund. Is there anything wrong with this thinking?
I also think you'd be better off in the PIMCO funds.... which give you the 100% equity pls roughly 100% bonds.

The LETF strategy only makes sense to me if you are wanting to go over 100% equity.

perplexed
Posts: 57
Joined: Thu Feb 02, 2012 4:32 pm

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

Post by perplexed » Mon Sep 09, 2019 10:44 am

Hydromod wrote:
Mon Sep 09, 2019 8:37 am
dspencer wrote:
Mon Sep 09, 2019 8:08 am
comeinvest wrote:
Mon Sep 09, 2019 2:56 am

To everybody: What's the proposed end game of Hedgefundie's strategy (or its variations) if U.S. rates follow the pattern of Japan and Europe? Would you still play this strategy if 10-year treasuries are at -1% and 30-year at -0.5% nominal, -2.5% and -2% real interest? If not, at what point would you pull the plug, or change the strategy?
I would not stay in this strategy at that point. I'm still trying to answer the last question.
I'm also still trying to answer the last question. I'm mulling over strategies for (i) backing down the leverage for some period of time, and (ii) possibly altering the mix of assets. For me, the key is to use the strategy as a tool to augment my overall portfolio when conditions are likely to be neutral to favorable.
Likewise. Sizeable gains in TMF took a jump down in the last one week. With the FED meet next week, I am unclear what message they are trying relay.

comeinvest
Posts: 284
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest » Mon Sep 09, 2019 5:49 pm

privatefarmer wrote:
Mon Sep 09, 2019 4:41 am
comeinvest wrote:
Mon Sep 09, 2019 2:56 am
NMBob wrote:
Sun Sep 08, 2019 7:41 pm
picking the gfc when people even feared government backed bonds for a while is a perhaps weakness, collapse of bond market and financial system....to this portfolio as discussed in the miles long thread. it is not the only time the market went down in history.

biggest annual stock drops since 1978

2002 - us stock down 20.96 long t's up 16.67%
2008 - us stock down 37.04 long ts up 22.51 %
To everybody: What's the proposed end game of Hedgefundie's strategy (or its variations) if U.S. rates follow the pattern of Japan and Europe? Would you still play this strategy if 10-year treasuries are at -1% and 30-year at -0.5% nominal, -2.5% and -2% real interest? If not, at what point would you pull the plug, or change the strategy?
- in that scenario borrowing costs would also be lower, decreasing the cost of leverage
- in that scenario, other assets should also have lower expected returns (ie equities), would it make more sense to then be only exposed to equity risk?
- is there really that much of a difference between making next to nothing in LTT yields compared to -0.5%?
- the market has already priced all this in. The reality is, IF LTTs went negative it would only be because the market has priced them so. Either out of due to fears of deflation, equity market crash, etc. or because the market expects future drops in yield resulting in price appreciation.
- if the market drove up equity P/E ratios to say 60+, would you no longer own equities?
I don't mean to refute your arguments, which in general may be very valid.
However, I'm taking an opposing view for the sake of a fruitful discussion.
- "borrowing costs would also be lower" - Borrowing cost are indeed ca. -0.5% in Europe. However those are current rates, while with long term BUNDs you are locked in for 30 years. Also, assuming that there is a lower limit for interest rates as people have the alternative of hoarding cash at home or in safe deposit boxes, it appears that in the *best* case, there would be no leverage for 30 years.
- "other assets should also have lower expected returns (ie equities)" - According to most forecasts including Vanguard's and Research Affiliates' which I respect a lot, real expected equity returns in Europe and most of the rest of the world are currently ca. 5-7%.
- "is there really that much of a difference..." - Yields in Europe are ca. -0.5% nominal, -2% real (1.5% inflation expectations). I think this number would exceed the expected return from the rebalancing effect.
- "the market has already priced all this in" - The market has many segments and interest groups, with different degrees of risk aversiveness, and different investment horizons, including government sponsored and regulated entities. What's good for somebody may not be good for somebody else.
- "if the market drove up equity P/E ratios to say 60+, would you no longer own equities?" - I might stay invested as nobody knows if expected returns would stay low forever; but given the historic example of Japan, I would probably either move into bonds (depending on bond yields), or at least eliminate leverage.

comeinvest
Posts: 284
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest » Mon Sep 09, 2019 6:05 pm

privatefarmer wrote:
Mon Sep 09, 2019 8:28 am
dspencer wrote:
Mon Sep 09, 2019 8:08 am
comeinvest wrote:
Mon Sep 09, 2019 2:56 am

To everybody: What's the proposed end game of Hedgefundie's strategy (or its variations) if U.S. rates follow the pattern of Japan and Europe? Would you still play this strategy if 10-year treasuries are at -1% and 30-year at -0.5% nominal, -2.5% and -2% real interest? If not, at what point would you pull the plug, or change the strategy?
I would not stay in this strategy at that point. I'm still trying to answer the last question.
this kind of sentiment is exactly what keeps the market "efficient". the only way yields would go negative is if the MARKET drove them negative. if the collective market of investors does not feel that negative yielding long-term debt is a worthwhile investment then they won't ever go negative.
Let me give you a thought experiment:
For the sake of simplicity of the argument and given the complexity of the stock market, imagine a hypothetical world, where there was only one asset class (bonds), that historically had real returns of 3%, but based on current yields, future returns are -2%. Cash yields are 0%, for the sake of this experiment. (Or cash could be -2%, which wouldn't change the essence of this thought experiment.)
You are a smart guy, and in the past you leveraged this asset class and achieved handsome returns over the decades and accumulated a nice retirement portfolio.
For the future, market participants expect population decline, terrorist activity, decimation of the rain forest, socialist movements, and possibly nuclear wars. Therefore consumers decided that dollars are worth more now than in the future. Likewise, businesses no longer want to borrow dollars to invest for this precarious future.
Would you still leverage your portfolio because "the market drove rates negative", and "the collective market of investors feels that negative yielding long-term debt is a worthwhile investment", thus accelerating the decimation of your portfolio?
(Hint: What if nobody thinks that this is a worthwhile investment, but there is just no or not enough counterparty demand for borrowed dollars? What if investors are no longer "entitled" to positive returns, which are not a law of nature, last time I checked?)

firebirdparts
Posts: 185
Joined: Thu Jun 13, 2019 4:21 pm

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

Post by firebirdparts » Tue Sep 10, 2019 10:04 am

Is this a deflationary scenario where people fear inflation? That's what it sounds like.

Anyway, I like thought experiments, so I will play anyway. I cannot go to the place (mentally) where I know what future returns will be. If I can borrow money at -2%, I would borrow it. However, I presume that only the government can do that. If the demand for capital is really that bad, then no I would not invest at all.

So I am saying based on what is given, you would not have a portfolio, but borrowing itself and holding cash offers some returns.
Last edited by firebirdparts on Tue Sep 10, 2019 9:29 pm, edited 1 time in total.

rascott
Posts: 702
Joined: Wed Apr 15, 2015 10:53 am

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

Post by rascott » Tue Sep 10, 2019 11:29 am

MotoTrojan wrote:
Sun Sep 08, 2019 7:28 pm
MinhN wrote:
Sun Sep 08, 2019 5:56 pm
I am considering using a modified version of this. With 1/3 in UPRO and 2/3 in bonds. It's like going 100% in equity with the added bonus of having extra bond income to offset the expense of running a leveraged fund. Is there anything wrong with this thinking?
I would not suggest holding TMF for the bond side but would instead use EDV. This is what I am using however I am using 43% UPRO and 57% EDV. EDV is not a leveraged product but is a Vanguard ETF that invests in long-term strips, essentially 20-30 year treasuries. Works just like a 20 year fund, but the duration is about 1.7x that of a 20 year fund.

1/3 UPRO and 2/3 EDV would've given you results very similar to the PSLDX fund that hedgefundie mentioned to you. My 43/57 variant performed a bit better and had drawdown similar to that of the unleveraged S&P500 and PSLDX (due to it's extra credit risk).

A 2/3 holding in TMF will get slaughtered if rates rise OR stay where they are at due to volatility decay; no such problem with EDV. The only case for using TMF is if you have a lot more equity exposure to offset.

I'm sure I've missed it in these long threads... but how did you end up at this combo (and the such exact %). Like you, I've been hesitant since the beginning of 3x leverage on LTTs (even though I'm already in the 40/60 setup for a small piece).

The idea of using STT futures to make somewhat of a synthetic long bond duration (with much less term risk) is something I'm playing around with.

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Tue Sep 10, 2019 11:39 am

rascott wrote:
Tue Sep 10, 2019 11:29 am
MotoTrojan wrote:
Sun Sep 08, 2019 7:28 pm
MinhN wrote:
Sun Sep 08, 2019 5:56 pm
I am considering using a modified version of this. With 1/3 in UPRO and 2/3 in bonds. It's like going 100% in equity with the added bonus of having extra bond income to offset the expense of running a leveraged fund. Is there anything wrong with this thinking?
I would not suggest holding TMF for the bond side but would instead use EDV. This is what I am using however I am using 43% UPRO and 57% EDV. EDV is not a leveraged product but is a Vanguard ETF that invests in long-term strips, essentially 20-30 year treasuries. Works just like a 20 year fund, but the duration is about 1.7x that of a 20 year fund.

1/3 UPRO and 2/3 EDV would've given you results very similar to the PSLDX fund that hedgefundie mentioned to you. My 43/57 variant performed a bit better and had drawdown similar to that of the unleveraged S&P500 and PSLDX (due to it's extra credit risk).

A 2/3 holding in TMF will get slaughtered if rates rise OR stay where they are at due to volatility decay; no such problem with EDV. The only case for using TMF is if you have a lot more equity exposure to offset.

I'm sure I've missed it in these long threads... but how did you end up at this combo (and the such exact %). Like you, I've been hesitant since the beginning of 3x leverage on LTTs (even though I'm already in the 40/60 setup for a small piece).

The idea of using STT futures to make somewhat of a synthetic long bond duration (with much less term risk) is something I'm playing around with.
It is roughly the same ratio of equity to duration as the 55/45 UPRO/TMF variant (I think I used volatility as a proxy for duration when comparing TMF & EDV using Simba). I also just played around with the Simba sheet in general and this balance felt right. I wanted a touch more equity/expected-return than the Pimco products too.

I don't think there is such a thing as "much less term risk" unless you have less effective duration. Some have shown that STT have better Sharpe ratio than LTT but I haven't seen enough evidence to suggest it is worth it unless you need more leverage/duration than you can achieve with EDV.

rascott
Posts: 702
Joined: Wed Apr 15, 2015 10:53 am

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

Post by rascott » Tue Sep 10, 2019 4:14 pm

MotoTrojan wrote:
Tue Sep 10, 2019 11:39 am
rascott wrote:
Tue Sep 10, 2019 11:29 am
MotoTrojan wrote:
Sun Sep 08, 2019 7:28 pm
MinhN wrote:
Sun Sep 08, 2019 5:56 pm
I am considering using a modified version of this. With 1/3 in UPRO and 2/3 in bonds. It's like going 100% in equity with the added bonus of having extra bond income to offset the expense of running a leveraged fund. Is there anything wrong with this thinking?
I would not suggest holding TMF for the bond side but would instead use EDV. This is what I am using however I am using 43% UPRO and 57% EDV. EDV is not a leveraged product but is a Vanguard ETF that invests in long-term strips, essentially 20-30 year treasuries. Works just like a 20 year fund, but the duration is about 1.7x that of a 20 year fund.

1/3 UPRO and 2/3 EDV would've given you results very similar to the PSLDX fund that hedgefundie mentioned to you. My 43/57 variant performed a bit better and had drawdown similar to that of the unleveraged S&P500 and PSLDX (due to it's extra credit risk).

A 2/3 holding in TMF will get slaughtered if rates rise OR stay where they are at due to volatility decay; no such problem with EDV. The only case for using TMF is if you have a lot more equity exposure to offset.

I'm sure I've missed it in these long threads... but how did you end up at this combo (and the such exact %). Like you, I've been hesitant since the beginning of 3x leverage on LTTs (even though I'm already in the 40/60 setup for a small piece).

The idea of using STT futures to make somewhat of a synthetic long bond duration (with much less term risk) is something I'm playing around with.
It is roughly the same ratio of equity to duration as the 55/45 UPRO/TMF variant (I think I used volatility as a proxy for duration when comparing TMF & EDV using Simba). I also just played around with the Simba sheet in general and this balance felt right. I wanted a touch more equity/expected-return than the Pimco products too.

I don't think there is such a thing as "much less term risk" unless you have less effective duration. Some have shown that STT have better Sharpe ratio than LTT but I haven't seen enough evidence to suggest it is worth it unless you need more leverage/duration than you can achieve with EDV.

I could have the terminology wrong... but looking back over the last 40 years or so.....STTs leveraged 6:1 had a CAGR 2.5% higher than a regular LTT fund .... and a lower std deviation.

And their negative correlation to the stock market was identical.

I don't know why that is or how it can be explained.

Anyway, I've started paper trading the 2yr futures note on TDA/TOS just to get a feel for it.

robertmcd
Posts: 511
Joined: Tue Aug 09, 2016 9:06 am

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

Post by robertmcd » Tue Sep 10, 2019 4:18 pm

^ shorter end of the curve tends to be steeper. Investors like pensions and insurance have to go out further for liability matching. Short term treasuries don't have a very high return without leverage, and many pensions/endowments/funds are not allowed to use leverage. Short term rates drop more than long term rates in time of crisis due to Fed policy focusing on lowering the short end of the curve the most, then letting the term premium bring the long end down.

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Tue Sep 10, 2019 5:27 pm

rascott wrote:
Tue Sep 10, 2019 4:14 pm


I could have the terminology wrong... but looking back over the last 40 years or so.....STTs leveraged 6:1 had a CAGR 2.5% higher than a regular LTT fund .... and a lower std deviation.

And their negative correlation to the stock market was identical.

I don't know why that is or how it can be explained.

Anyway, I've started paper trading the 2yr futures note on TDA/TOS just to get a feel for it.
How did you come up with the 6:1? How are you simulating that leverage? If that is truly the same effective-duration then that is a massive outperformance, but it doesn't seem like it is. Unleveraged STTs have the same correlation as leveraged ones and thus can have the same correlation as LTT, but that doesn't mean they'll respond the same to a rate movement.

Given that the std deviation was lower I would wager the duration was too; only way I can think of for the correlation to be identical but std dev off.

comeinvest
Posts: 284
Joined: Mon Mar 12, 2012 6:57 pm

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

Post by comeinvest » Tue Sep 10, 2019 5:37 pm

firebirdparts wrote:
Tue Sep 10, 2019 10:04 am
Is this a deflationary scenario where people fear inflation? That's what it sounds like.

Anyway, I like thought experiments, so I will play anyway. I cannot go to the place where I think the know what future returns will be. If I can borrow money at -2%, I would borrow it. However, I presume that only the government can do that. If the demand for capital is really that bad, then no I would not invest at all.

So I am saying based on what is given, you would not have a portfolio, but borrowing itself and holding cash offers some returns.
No, I think my scenario was a world with a continuing slightly inflationary scenario with permanently negative real risk-free returns. The numbers in the thought experiment were real returns, and are approximately the past and current real yields in Europe. The proponents of "staying the course" with the leveraged strategy often seem to assume that "the market" drove down the yields, and therefore "the market" assumes that this is justified because of either a sustained long-term period of deflation, or yields will go even lower (with no limit?), resulting in positive returns from yield change. I also see a sense of entitlement to positive returns on capital, especially risk-free returns. I don't have too much confidence that either will happen, and I am not aware of a law of nature, or other compelling evidence, that says returns have to be positive; therefore I was instead laying out a scenario of permanently negative risk-free returns. This was in response to a reply of a poster who was willing to continue the leveraged strategy with current European bond yields.
Last edited by comeinvest on Tue Sep 10, 2019 5:49 pm, edited 3 times in total.

rascott
Posts: 702
Joined: Wed Apr 15, 2015 10:53 am

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

Post by rascott » Tue Sep 10, 2019 5:46 pm

MotoTrojan wrote:
Tue Sep 10, 2019 5:27 pm
rascott wrote:
Tue Sep 10, 2019 4:14 pm


I could have the terminology wrong... but looking back over the last 40 years or so.....STTs leveraged 6:1 had a CAGR 2.5% higher than a regular LTT fund .... and a lower std deviation.

And their negative correlation to the stock market was identical.

I don't know why that is or how it can be explained.

Anyway, I've started paper trading the 2yr futures note on TDA/TOS just to get a feel for it.
How did you come up with the 6:1? How are you simulating that leverage? If that is truly the same effective-duration then that is a massive outperformance, but it doesn't seem like it is. Unleveraged STTs have the same correlation as leveraged ones and thus can have the same correlation as LTT, but that doesn't mean they'll respond the same to a rate movement.

Given that the std deviation was lower I would wager the duration was too; only way I can think of for the correlation to be identical but std dev off.
6 to 1 was just looking back since early 80s.... and they were both roughly -0.13 correlated to the market. You'd need to go 7-8:1 over that time frame to get the same volatility as a LTT fund.

I was just using PV and negative cash for the leverage situation.

It's varied based upon time periods..... since the GFC and ultra low rates, you'd need to go to something like 15x STTs to match the std deviation of LTTs. Still early in my exploring, but have yet to find a time period where leveraged STTs didn't beat LTTs once std deviation is equalized.

I "bought" a few /ZT Dec futures in the paper account today at 107'210.... just to get an idea what it's like to trade these. 3 contracts would be $600k exposure..... took all of about $2100 in liquidity :P

2 year is at 1.68%. If it's unchanged by Dec, I'd fully expect to lose a bit of fake money with the inverted curve. "Bought" $50k of SPY as the 12:1 ratio seems about right for this low rate environment. That's probably actually low if we are looking for risk parity.

robertmcd
Posts: 511
Joined: Tue Aug 09, 2016 9:06 am

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

Post by robertmcd » Wed Sep 11, 2019 12:09 pm

Can some explain to me what the corresponding VTI/EDV ratio would be to match the 55/45 upro/tmf

60/40 VTI/EDV is risk parity
40/60 UPRO/TMF is risk parity

New allocation based on historical data :

55/45 UPRO/TMF

?/? VTI/EDV

drzzzzz
Posts: 420
Joined: Sat Sep 22, 2012 9:56 pm

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

Post by drzzzzz » Wed Sep 11, 2019 1:30 pm

Can someone clarify something for me - if you are rebalancing quarterly are you rebalancing to 40/60 upro/tmf or rebalancing using portfolio visualizer to the market timing percentages in that monthly worksheet? For example as of September 10 it says 49/51 upro/tmf. While August appears to have been 55/45 upro/tmf and beginning of September 41/59 upro/tmf thanks

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Wed Sep 11, 2019 1:52 pm

drzzzzz wrote:
Wed Sep 11, 2019 1:30 pm
Can someone clarify something for me - if you are rebalancing quarterly are you rebalancing to 40/60 upro/tmf or rebalancing using portfolio visualizer to the market timing percentages in that monthly worksheet? For example as of September 10 it says 49/51 upro/tmf. While August appears to have been 55/45 upro/tmf and beginning of September 41/59 upro/tmf thanks
Everyone has their own strategy. HF is doing 55/45 quarterly, no market timing. Some are doing 1-3 month look-back risk parity. Others ignore TMF and are doing target volatility on UPRO (I know someone going as high as 25% target vol, which is often at 100% UPRO when markets are calm).

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Wed Sep 11, 2019 1:58 pm

robertmcd wrote:
Wed Sep 11, 2019 12:09 pm
Can some explain to me what the corresponding VTI/EDV ratio would be to match the 55/45 upro/tmf

60/40 VTI/EDV is risk parity
40/60 UPRO/TMF is risk parity

New allocation based on historical data :

55/45 UPRO/TMF

?/? VTI/EDV
EDV has an average duration of 24 years while I believe TMF is closer to 53.2 (3x TLT's), so TMF is about 2.22x as potent as EDV. You could use excel to calculate and maintain the same ratio of equity exposure to duration. I would use VOO, not VTI. Why do you want this though?

robertmcd
Posts: 511
Joined: Tue Aug 09, 2016 9:06 am

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

Post by robertmcd » Wed Sep 11, 2019 2:41 pm

MotoTrojan wrote:
Wed Sep 11, 2019 1:58 pm
robertmcd wrote:
Wed Sep 11, 2019 12:09 pm
Can some explain to me what the corresponding VTI/EDV ratio would be to match the 55/45 upro/tmf

60/40 VTI/EDV is risk parity
40/60 UPRO/TMF is risk parity

New allocation based on historical data :

55/45 UPRO/TMF

?/? VTI/EDV
EDV has an average duration of 24 years while I believe TMF is closer to 53.2 (3x TLT's), so TMF is about 2.22x as potent as EDV. You could use excel to calculate and maintain the same ratio of equity exposure to duration. I would use VOO, not VTI. Why do you want this though?
I want to be able to go back and forth between equity overweight and duration overweight vs. the long run risk parity but I want to get a better way to quantify my exposure. I liked how hedgefundie did it based on historical data showing equity outperformance after a large drop in rates, and rates so low. I don't see anything wrong with using VTI vs S&P 500.

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Wed Sep 11, 2019 3:02 pm

robertmcd wrote:
Wed Sep 11, 2019 2:41 pm
MotoTrojan wrote:
Wed Sep 11, 2019 1:58 pm
robertmcd wrote:
Wed Sep 11, 2019 12:09 pm
Can some explain to me what the corresponding VTI/EDV ratio would be to match the 55/45 upro/tmf

60/40 VTI/EDV is risk parity
40/60 UPRO/TMF is risk parity

New allocation based on historical data :

55/45 UPRO/TMF

?/? VTI/EDV
EDV has an average duration of 24 years while I believe TMF is closer to 53.2 (3x TLT's), so TMF is about 2.22x as potent as EDV. You could use excel to calculate and maintain the same ratio of equity exposure to duration. I would use VOO, not VTI. Why do you want this though?
I want to be able to go back and forth between equity overweight and duration overweight vs. the long run risk parity but I want to get a better way to quantify my exposure. I liked how hedgefundie did it based on historical data showing equity outperformance after a large drop in rates, and rates so low. I don't see anything wrong with using VTI vs S&P 500.
Not following; 55/45 UPRO/TMF is intended to have the same weight to equity vs. duration as 43/57 UPRO/EDV, and it sounds like you want to know what would be the equivalent with VTI/EDV as well. The only thing changing between these is total leverage; no change in weight towards equity or rates.

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Wed Sep 11, 2019 3:10 pm

robertmcd wrote:
Wed Sep 11, 2019 12:09 pm
Can some explain to me what the corresponding VTI/EDV ratio would be to match the 55/45 upro/tmf

60/40 VTI/EDV is risk parity
40/60 UPRO/TMF is risk parity

New allocation based on historical data :

55/45 UPRO/TMF

?/? VTI/EDV
To answer your original question though it looks like 70% S&P500 (VTI should be similar) 30% EDV would be a similar ratio, which is interesting to me because 70/30 is a pretty normal equity/stock allocation for the long-haul (EDV is a special beast of course...).

For reference I calculated this using Simba data from 1955-1982 and using volatility rather than effective-duration directly (similar to how risk-parity is calculated with volatility).

Eqn is simple: (% equity X % volatility)/(% bond X % volatility). For 55/45 UPRO/TMF this was about 1.64, so I just adjusted the allocation of the other asset combo to achieve this ratio.

From this you can see that 43/57 UPRO/EDV is about 22% less leverage than hedgefundie's strategy, and the 70/30 VTI/EDV would be about 58% less.

robertmcd
Posts: 511
Joined: Tue Aug 09, 2016 9:06 am

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

Post by robertmcd » Wed Sep 11, 2019 3:36 pm

MotoTrojan wrote:
Wed Sep 11, 2019 3:10 pm
robertmcd wrote:
Wed Sep 11, 2019 12:09 pm
Can some explain to me what the corresponding VTI/EDV ratio would be to match the 55/45 upro/tmf

60/40 VTI/EDV is risk parity
40/60 UPRO/TMF is risk parity

New allocation based on historical data :

55/45 UPRO/TMF

?/? VTI/EDV
To answer your original question though it looks like 70% S&P500 (VTI should be similar) 30% EDV would be a similar ratio, which is interesting to me because 70/30 is a pretty normal equity/stock allocation for the long-haul (EDV is a special beast of course...).

For reference I calculated this using Simba data from 1955-1982 and using volatility rather than effective-duration directly (similar to how risk-parity is calculated with volatility).

Eqn is simple: (% equity X % volatility)/(% bond X % volatility). For 55/45 UPRO/TMF this was about 1.64, so I just adjusted the allocation of the other asset combo to achieve this ratio.

From this you can see that 43/57 UPRO/EDV is about 22% less leverage than hedgefundie's strategy, and the 70/30 VTI/EDV would be about 58% less.
Thank you this is what I was looking for. I just wanted to see the math so I could follow along with hedgefundie's changes and make changes of my own if I was comfortable with it.

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Wed Sep 11, 2019 4:07 pm

robertmcd wrote:
Wed Sep 11, 2019 3:36 pm
MotoTrojan wrote:
Wed Sep 11, 2019 3:10 pm
robertmcd wrote:
Wed Sep 11, 2019 12:09 pm
Can some explain to me what the corresponding VTI/EDV ratio would be to match the 55/45 upro/tmf

60/40 VTI/EDV is risk parity
40/60 UPRO/TMF is risk parity

New allocation based on historical data :

55/45 UPRO/TMF

?/? VTI/EDV
To answer your original question though it looks like 70% S&P500 (VTI should be similar) 30% EDV would be a similar ratio, which is interesting to me because 70/30 is a pretty normal equity/stock allocation for the long-haul (EDV is a special beast of course...).

For reference I calculated this using Simba data from 1955-1982 and using volatility rather than effective-duration directly (similar to how risk-parity is calculated with volatility).

Eqn is simple: (% equity X % volatility)/(% bond X % volatility). For 55/45 UPRO/TMF this was about 1.64, so I just adjusted the allocation of the other asset combo to achieve this ratio.

From this you can see that 43/57 UPRO/EDV is about 22% less leverage than hedgefundie's strategy, and the 70/30 VTI/EDV would be about 58% less.
Thank you this is what I was looking for. I just wanted to see the math so I could follow along with hedgefundie's changes and make changes of my own if I was comfortable with it.
Right on! To be clear I don't think hedgefundie used this at all, he just picked 55/45 because it felt right, but I wanted to remove leverage entirely on the bond side (eliminate volatility decay there in a flat rate environment), and maintain the same ratio because I liked the back-test, and I found it an interesting comparison (basically the same portfolio with 22% less leverage).

Topic Author
HEDGEFUNDIE
Posts: 3286
Joined: Sun Oct 22, 2017 2:06 pm

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

Post by HEDGEFUNDIE » Wed Sep 11, 2019 5:35 pm

MotoTrojan wrote:
Wed Sep 11, 2019 4:07 pm
robertmcd wrote:
Wed Sep 11, 2019 3:36 pm
MotoTrojan wrote:
Wed Sep 11, 2019 3:10 pm
robertmcd wrote:
Wed Sep 11, 2019 12:09 pm
Can some explain to me what the corresponding VTI/EDV ratio would be to match the 55/45 upro/tmf

60/40 VTI/EDV is risk parity
40/60 UPRO/TMF is risk parity

New allocation based on historical data :

55/45 UPRO/TMF

?/? VTI/EDV
To answer your original question though it looks like 70% S&P500 (VTI should be similar) 30% EDV would be a similar ratio, which is interesting to me because 70/30 is a pretty normal equity/stock allocation for the long-haul (EDV is a special beast of course...).

For reference I calculated this using Simba data from 1955-1982 and using volatility rather than effective-duration directly (similar to how risk-parity is calculated with volatility).

Eqn is simple: (% equity X % volatility)/(% bond X % volatility). For 55/45 UPRO/TMF this was about 1.64, so I just adjusted the allocation of the other asset combo to achieve this ratio.

From this you can see that 43/57 UPRO/EDV is about 22% less leverage than hedgefundie's strategy, and the 70/30 VTI/EDV would be about 58% less.
Thank you this is what I was looking for. I just wanted to see the math so I could follow along with hedgefundie's changes and make changes of my own if I was comfortable with it.
Right on! To be clear I don't think hedgefundie used this at all, he just picked 55/45 because it felt right, but I wanted to remove leverage entirely on the bond side (eliminate volatility decay there in a flat rate environment), and maintain the same ratio because I liked the back-test, and I found it an interesting comparison (basically the same portfolio with 22% less leverage).
I will ignore the "because it felt right" swipe...

...and point out that what you have is not "basically the same portfolio". If stocks crash and long rates collapse (perhaps into negative territory) EDV will not blunt the pain nearly as much as TMF.

rascott
Posts: 702
Joined: Wed Apr 15, 2015 10:53 am

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

Post by rascott » Wed Sep 11, 2019 9:07 pm

HEDGEFUNDIE wrote:
Wed Sep 11, 2019 5:35 pm
MotoTrojan wrote:
Wed Sep 11, 2019 4:07 pm
robertmcd wrote:
Wed Sep 11, 2019 3:36 pm
MotoTrojan wrote:
Wed Sep 11, 2019 3:10 pm
robertmcd wrote:
Wed Sep 11, 2019 12:09 pm
Can some explain to me what the corresponding VTI/EDV ratio would be to match the 55/45 upro/tmf

60/40 VTI/EDV is risk parity
40/60 UPRO/TMF is risk parity

New allocation based on historical data :

55/45 UPRO/TMF

?/? VTI/EDV
To answer your original question though it looks like 70% S&P500 (VTI should be similar) 30% EDV would be a similar ratio, which is interesting to me because 70/30 is a pretty normal equity/stock allocation for the long-haul (EDV is a special beast of course...).

For reference I calculated this using Simba data from 1955-1982 and using volatility rather than effective-duration directly (similar to how risk-parity is calculated with volatility).

Eqn is simple: (% equity X % volatility)/(% bond X % volatility). For 55/45 UPRO/TMF this was about 1.64, so I just adjusted the allocation of the other asset combo to achieve this ratio.

From this you can see that 43/57 UPRO/EDV is about 22% less leverage than hedgefundie's strategy, and the 70/30 VTI/EDV would be about 58% less.
Thank you this is what I was looking for. I just wanted to see the math so I could follow along with hedgefundie's changes and make changes of my own if I was comfortable with it.
Right on! To be clear I don't think hedgefundie used this at all, he just picked 55/45 because it felt right, but I wanted to remove leverage entirely on the bond side (eliminate volatility decay there in a flat rate environment), and maintain the same ratio because I liked the back-test, and I found it an interesting comparison (basically the same portfolio with 22% less leverage).
I will ignore the "because it felt right" swipe...

...and point out that what you have is not "basically the same portfolio". If stocks crash and long rates collapse (perhaps into negative territory) EDV will not blunt the pain nearly as much as TMF.
What are your thoughts on the leveraged ST Treasuries.... via futures?

I've run every period I could, back to the 50s (thanks EfficientInvestor)... and have found leveraged short terms.... varying anywhere from 6x to nearly 20x, depending upon the overall Fed Funds rate environment.... has 1) higher Sharpe when combined with leveraged equities and subsequently 2) just as much cushion (or more) as equitable (same vol) level of leveraged LTTs.

It's not a set and forget system, as you've got to roll futures every quarter (which seems as much work as the M1 rebalance button). And you would need to do some basic calculations to get your ratios right each qtr between futures and actual holdings. In other words might take an hour each qtr.... instead of 30 secs. And at least save the mgmt fee on the bonds. As you know, the LETF on the LTTs has always bothered me.

That said, you could still use UPRO to whatever element you wanted for the equity side. And possibly eliminate the decay concern of TMF. I think you'd need to be levered 20:1 (in current rate environment) or so to equateTMF to levered STTs... but that's easily doable with futures. Without mgmt fees and no decay issue.

I'm just spit balling here right now, and am just scratching the surface, myself. When I have time at a regular PC I'll try to post in more detail of what I'm seeing/ thinking.

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Wed Sep 11, 2019 10:25 pm

HEDGEFUNDIE wrote:
Wed Sep 11, 2019 5:35 pm
MotoTrojan wrote:
Wed Sep 11, 2019 4:07 pm


Right on! To be clear I don't think hedgefundie used this at all, he just picked 55/45 because it felt right, but I wanted to remove leverage entirely on the bond side (eliminate volatility decay there in a flat rate environment), and maintain the same ratio because I liked the back-test, and I found it an interesting comparison (basically the same portfolio with 22% less leverage).
I will ignore the "because it felt right" swipe...

...and point out that what you have is not "basically the same portfolio". If stocks crash and long rates collapse (perhaps into negative territory) EDV will not blunt the pain nearly as much as TMF.


I think you are letting what you thought was a swipe get in the way of understanding what I have done, because what you pointed out is incorrect. I have the exact same ratio of equity exposure to bond duration as your portfolio, so if stocks crash and long rates collapse my equity portion will not need as as much pain blunting, and will actually be equally protected, just with less leverage, so theoretically it should actually do better in a downturn, just not as well when the portfolio positive.

I recall I was one of the 1st people to ask you if you had a quantitative rationale for 55/45 and you agreed with my educated guess that you changed it because equities had infinite upside potential and bonds (due to low rates) had a potentially finite amount of upside; there was no quantitative reason for 55/45 (as there was for 40/60), but it "felt right". Where am I wrong there?

Topic Author
HEDGEFUNDIE
Posts: 3286
Joined: Sun Oct 22, 2017 2:06 pm

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

Post by HEDGEFUNDIE » Wed Sep 11, 2019 10:42 pm

MotoTrojan wrote:
Wed Sep 11, 2019 10:25 pm
HEDGEFUNDIE wrote:
Wed Sep 11, 2019 5:35 pm
MotoTrojan wrote:
Wed Sep 11, 2019 4:07 pm


Right on! To be clear I don't think hedgefundie used this at all, he just picked 55/45 because it felt right, but I wanted to remove leverage entirely on the bond side (eliminate volatility decay there in a flat rate environment), and maintain the same ratio because I liked the back-test, and I found it an interesting comparison (basically the same portfolio with 22% less leverage).
I will ignore the "because it felt right" swipe...

...and point out that what you have is not "basically the same portfolio". If stocks crash and long rates collapse (perhaps into negative territory) EDV will not blunt the pain nearly as much as TMF.


I think you are letting what you thought was a swipe get in the way of understanding what I have done, because what you pointed out is incorrect. I have the exact same ratio of equity exposure to bond duration as your portfolio, so if stocks crash and long rates collapse my equity portion will not need as as much pain blunting, and will actually be equally protected, just with less leverage, so theoretically it should actually do better in a downturn, just not as well when the portfolio positive.

I recall I was one of the 1st people to ask you if you had a quantitative rationale for 55/45 and you agreed with my educated guess that you changed it because equities had infinite upside potential and bonds (due to low rates) had a potentially finite amount of upside; there was no quantitative reason for 55/45 (as there was for 40/60), but it "felt right". Where am I wrong there?
Re-run risk parity AA from the first time 30 year rates were this low (July 2012) to present and you get an AA of 51/49. Plus an extra 4% for the Adventure 😉

rascott
Posts: 702
Joined: Wed Apr 15, 2015 10:53 am

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

Post by rascott » Wed Sep 11, 2019 11:51 pm

HEDGEFUNDIE wrote:
Wed Sep 11, 2019 10:42 pm
MotoTrojan wrote:
Wed Sep 11, 2019 10:25 pm
HEDGEFUNDIE wrote:
Wed Sep 11, 2019 5:35 pm
MotoTrojan wrote:
Wed Sep 11, 2019 4:07 pm


Right on! To be clear I don't think hedgefundie used this at all, he just picked 55/45 because it felt right, but I wanted to remove leverage entirely on the bond side (eliminate volatility decay there in a flat rate environment), and maintain the same ratio because I liked the back-test, and I found it an interesting comparison (basically the same portfolio with 22% less leverage).
I will ignore the "because it felt right" swipe...

...and point out that what you have is not "basically the same portfolio". If stocks crash and long rates collapse (perhaps into negative territory) EDV will not blunt the pain nearly as much as TMF.


I think you are letting what you thought was a swipe get in the way of understanding what I have done, because what you pointed out is incorrect. I have the exact same ratio of equity exposure to bond duration as your portfolio, so if stocks crash and long rates collapse my equity portion will not need as as much pain blunting, and will actually be equally protected, just with less leverage, so theoretically it should actually do better in a downturn, just not as well when the portfolio positive.

I recall I was one of the 1st people to ask you if you had a quantitative rationale for 55/45 and you agreed with my educated guess that you changed it because equities had infinite upside potential and bonds (due to low rates) had a potentially finite amount of upside; there was no quantitative reason for 55/45 (as there was for 40/60), but it "felt right". Where am I wrong there?
Re-run risk parity AA from the first time 30 year rates were this low (July 2012) to present and you get an AA of 51/49. Plus an extra 4% for the Adventure 😉
Not sure I understand this. Are you saying that equities are expected to be less volatile in a low rate environment than historical average?

Looking at only the last 7 years doesn't help model.

Topic Author
HEDGEFUNDIE
Posts: 3286
Joined: Sun Oct 22, 2017 2:06 pm

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

Post by HEDGEFUNDIE » Thu Sep 12, 2019 12:15 am

rascott wrote:
Wed Sep 11, 2019 11:51 pm
HEDGEFUNDIE wrote:
Wed Sep 11, 2019 10:42 pm
MotoTrojan wrote:
Wed Sep 11, 2019 10:25 pm
HEDGEFUNDIE wrote:
Wed Sep 11, 2019 5:35 pm
MotoTrojan wrote:
Wed Sep 11, 2019 4:07 pm


Right on! To be clear I don't think hedgefundie used this at all, he just picked 55/45 because it felt right, but I wanted to remove leverage entirely on the bond side (eliminate volatility decay there in a flat rate environment), and maintain the same ratio because I liked the back-test, and I found it an interesting comparison (basically the same portfolio with 22% less leverage).
I will ignore the "because it felt right" swipe...

...and point out that what you have is not "basically the same portfolio". If stocks crash and long rates collapse (perhaps into negative territory) EDV will not blunt the pain nearly as much as TMF.


I think you are letting what you thought was a swipe get in the way of understanding what I have done, because what you pointed out is incorrect. I have the exact same ratio of equity exposure to bond duration as your portfolio, so if stocks crash and long rates collapse my equity portion will not need as as much pain blunting, and will actually be equally protected, just with less leverage, so theoretically it should actually do better in a downturn, just not as well when the portfolio positive.

I recall I was one of the 1st people to ask you if you had a quantitative rationale for 55/45 and you agreed with my educated guess that you changed it because equities had infinite upside potential and bonds (due to low rates) had a potentially finite amount of upside; there was no quantitative reason for 55/45 (as there was for 40/60), but it "felt right". Where am I wrong there?
Re-run risk parity AA from the first time 30 year rates were this low (July 2012) to present and you get an AA of 51/49. Plus an extra 4% for the Adventure 😉
Not sure I understand this. Are you saying that equities are expected to be less volatile in a low rate environment than historical average?

Looking at only the last 7 years doesn't help model.
I’m saying that what is risk parity changes (obviously) and if I’m not going to do the adaptive strategy (which I’m not) then calculating risk parity starting with a time when rates were last this low is more defensible than when rates were 15%.

rascott
Posts: 702
Joined: Wed Apr 15, 2015 10:53 am

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

Post by rascott » Thu Sep 12, 2019 12:26 am

HEDGEFUNDIE wrote:
Thu Sep 12, 2019 12:15 am
rascott wrote:
Wed Sep 11, 2019 11:51 pm
HEDGEFUNDIE wrote:
Wed Sep 11, 2019 10:42 pm
MotoTrojan wrote:
Wed Sep 11, 2019 10:25 pm
HEDGEFUNDIE wrote:
Wed Sep 11, 2019 5:35 pm


I will ignore the "because it felt right" swipe...

...and point out that what you have is not "basically the same portfolio". If stocks crash and long rates collapse (perhaps into negative territory) EDV will not blunt the pain nearly as much as TMF.


I think you are letting what you thought was a swipe get in the way of understanding what I have done, because what you pointed out is incorrect. I have the exact same ratio of equity exposure to bond duration as your portfolio, so if stocks crash and long rates collapse my equity portion will not need as as much pain blunting, and will actually be equally protected, just with less leverage, so theoretically it should actually do better in a downturn, just not as well when the portfolio positive.

I recall I was one of the 1st people to ask you if you had a quantitative rationale for 55/45 and you agreed with my educated guess that you changed it because equities had infinite upside potential and bonds (due to low rates) had a potentially finite amount of upside; there was no quantitative reason for 55/45 (as there was for 40/60), but it "felt right". Where am I wrong there?
Re-run risk parity AA from the first time 30 year rates were this low (July 2012) to present and you get an AA of 51/49. Plus an extra 4% for the Adventure 😉
Not sure I understand this. Are you saying that equities are expected to be less volatile in a low rate environment than historical average?

Looking at only the last 7 years doesn't help model.
I’m saying that what is risk parity changes (obviously) and if I’m not going to do the adaptive strategy (which I’m not) then calculating risk parity starting with a time when rates were last this low is more defensible than when rates were 15%.
I'm in agreement, we want to lev equities with insurance. Insurace is more expensive in low rates.

The adventure is low risk lottery ticket for me. Others want a lower risk option for a larger position of their portfolio.

I'm targeting 20% of my total nut to leveraged plays. 10% PIMCO does it, 10% I do it. The 10% I do I want to be uber aggressive (which I feel is 1.5x-3x equities). In line with this.

dspencer
Posts: 195
Joined: Wed Jul 06, 2016 11:29 am

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

Post by dspencer » Fri Sep 13, 2019 8:36 am

MotoTrojan wrote:
Wed Sep 11, 2019 3:10 pm
robertmcd wrote:
Wed Sep 11, 2019 12:09 pm
Can some explain to me what the corresponding VTI/EDV ratio would be to match the 55/45 upro/tmf

60/40 VTI/EDV is risk parity
40/60 UPRO/TMF is risk parity

New allocation based on historical data :

55/45 UPRO/TMF

?/? VTI/EDV
To answer your original question though it looks like 70% S&P500 (VTI should be similar) 30% EDV would be a similar ratio, which is interesting to me because 70/30 is a pretty normal equity/stock allocation for the long-haul (EDV is a special beast of course...).

For reference I calculated this using Simba data from 1955-1982 and using volatility rather than effective-duration directly (similar to how risk-parity is calculated with volatility).

Eqn is simple: (% equity X % volatility)/(% bond X % volatility). For 55/45 UPRO/TMF this was about 1.64, so I just adjusted the allocation of the other asset combo to achieve this ratio.

From this you can see that 43/57 UPRO/EDV is about 22% less leverage than hedgefundie's strategy, and the 70/30 VTI/EDV would be about 58% less.
MotoTrojan,

Are you essentially considering EDV as a leveraged fund even though it is technically not leveraged? What are you considering it's effective leverage to be?

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Fri Sep 13, 2019 8:46 am

dspencer wrote:
Fri Sep 13, 2019 8:36 am
MotoTrojan wrote:
Wed Sep 11, 2019 3:10 pm
robertmcd wrote:
Wed Sep 11, 2019 12:09 pm
Can some explain to me what the corresponding VTI/EDV ratio would be to match the 55/45 upro/tmf

60/40 VTI/EDV is risk parity
40/60 UPRO/TMF is risk parity

New allocation based on historical data :

55/45 UPRO/TMF

?/? VTI/EDV
To answer your original question though it looks like 70% S&P500 (VTI should be similar) 30% EDV would be a similar ratio, which is interesting to me because 70/30 is a pretty normal equity/stock allocation for the long-haul (EDV is a special beast of course...).

For reference I calculated this using Simba data from 1955-1982 and using volatility rather than effective-duration directly (similar to how risk-parity is calculated with volatility).

Eqn is simple: (% equity X % volatility)/(% bond X % volatility). For 55/45 UPRO/TMF this was about 1.64, so I just adjusted the allocation of the other asset combo to achieve this ratio.

From this you can see that 43/57 UPRO/EDV is about 22% less leverage than hedgefundie's strategy, and the 70/30 VTI/EDV would be about 58% less.
MotoTrojan,

Are you essentially considering EDV as a leveraged fund even though it is technically not leveraged? What are you considering it's effective leverage to be?
Essentially yes, although the 22% calc is simply the reduction in UPRO exposure. On the equity side you are leveraging to boost returns, on the bond side what you are really doing is leveraging to boost the effective-duration. Looking purely at how much leverage as a % you are using on the bond side is somewhat flawed as 3x leverage on 20 year bonds is much different than 3x leverage on short-term treasuries. If we use 55/45 UPRO/TMF as the comparison point, 43/57 UPRO/EDV is a 22% reduction in equity exposure/leverage, with an effective-duration ratio held constant (again, I used volatility via Simba from 1955-2018 as a proxy for duration).

Regardless of how I got there, I like how this allocation behaves through back-testing via Simba as well as the data since VEDTX inception (2008 I believe). It gives me treasury exposure rather than corporates (such as Pimco StocksPLUS) and it edges up the equity exposure somewhat, which will boost the overall expected return while reducing the exposure the rates. Most importantly compared to the OPs strategy though is that it eliminates the path dependency on the bond side; if 20 year rates stay where they are today for the next 30 years, a portfolio holding unleveraged S&P500/LTT may still have a great Sharpe ratio, but the portfolio using TMF could get crushed due to volatility decay, where-as the EDV holding will providing some income and rebalancing bonus without said decay. Even if rates work their way down to 0% over 30 years, EDV could still very likely come out ahead with enough bond volatility.

If we (IMHO wrongly) assume that 55/45 UPRO/TMF will perform similarly moving forward as it has in the past 37 years (20% CAGR) then I am still going to have monster performance (17% CAGR), and even if things are closer to 1955-present I should still beat the S&P500 by 1-2% (while OP may be more like 2-3%), but there are several scenarios where I beat the S&P500 while the OP doesn't.

rascott
Posts: 702
Joined: Wed Apr 15, 2015 10:53 am

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

Post by rascott » Fri Sep 13, 2019 11:25 am

I've decided to eventually transition my lottery ticket play to the futures market. Eliminates the mgmt fee, also eliminates the volatility decay of TMF.

150% SP500
1500% Short-Term Treasuries

This isn't a static leverage level...in higher rate environments one would dial down the leverage of STTs to meet the target duration.

From 2010 (current rate world)

https://www.portfoliovisualizer.com/bac ... total3=100

From early 90s....using the long-run ratio of 6:1 STTs/Equities....shooting for risk parity:

https://www.portfoliovisualizer.com/bac ... total3=100

The higher Sharpe allows for a higher equity allocation:

https://www.portfoliovisualizer.com/bac ... total3=100
Last edited by rascott on Fri Sep 13, 2019 11:28 am, edited 1 time in total.

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Fri Sep 13, 2019 11:27 am

rascott wrote:
Fri Sep 13, 2019 11:25 am
I've decided to eventually transition my lottery ticket play to the futures market. Eliminates the mgmt fee, also eliminates the volatility decay of TMF.

150% SP500
1500% Short-Term Treasuries

This isn't a static leverage level...in higher rate environments one would dial down the leverage of STTs to meet the target duration.

From 2010 (current rate world)

https://www.portfoliovisualizer.com/bac ... total3=100

From early 90s....using the long-run ratio of 6:1 STTs/Equities.

https://www.portfoliovisualizer.com/bac ... total3=100
Why does duration change based on rate? That doesn't make sense.

rascott
Posts: 702
Joined: Wed Apr 15, 2015 10:53 am

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

Post by rascott » Fri Sep 13, 2019 11:31 am

MotoTrojan wrote:
Fri Sep 13, 2019 11:27 am
rascott wrote:
Fri Sep 13, 2019 11:25 am
I've decided to eventually transition my lottery ticket play to the futures market. Eliminates the mgmt fee, also eliminates the volatility decay of TMF.

150% SP500
1500% Short-Term Treasuries

This isn't a static leverage level...in higher rate environments one would dial down the leverage of STTs to meet the target duration.

From 2010 (current rate world)

https://www.portfoliovisualizer.com/bac ... total3=100

From early 90s....using the long-run ratio of 6:1 STTs/Equities.

https://www.portfoliovisualizer.com/bac ... total3=100
Why does duration change based on rate? That doesn't make sense.
Probably not the right terminology. But when rates are lower, you have to leverage STTs at a higher rate for them to equalize the volatility of LTTs. Maybe someone smarter than I can explain that.

SHY has a duration of 1.89 currently. If I wanted to equate that to TMF, I'd need to lever it 28X. 60% (original risk parity TMF allocation) of 28x leverage gets me at roughly 16.8....or 1680% STTs.

But in eras of higher rates, this level of leverage was way too high.

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Fri Sep 13, 2019 11:44 am

rascott wrote:
Fri Sep 13, 2019 11:31 am
MotoTrojan wrote:
Fri Sep 13, 2019 11:27 am
rascott wrote:
Fri Sep 13, 2019 11:25 am
I've decided to eventually transition my lottery ticket play to the futures market. Eliminates the mgmt fee, also eliminates the volatility decay of TMF.

150% SP500
1500% Short-Term Treasuries

This isn't a static leverage level...in higher rate environments one would dial down the leverage of STTs to meet the target duration.

From 2010 (current rate world)

https://www.portfoliovisualizer.com/bac ... total3=100

From early 90s....using the long-run ratio of 6:1 STTs/Equities.

https://www.portfoliovisualizer.com/bac ... total3=100
Why does duration change based on rate? That doesn't make sense.
Probably not the right terminology. But when rates are lower, you have to leverage STTs at a higher rate for them to equalize the volatility of LTTs. Maybe someone smarter than I can explain that.

SHY has a duration of 1.89 currently. If I wanted to equate that to TMF, I'd need to lever it 28X. 60% (original risk parity TMF allocation) of 28x leverage gets me at roughly 16.8....or 1680% STTs.

But in eras of higher rates, this level of leverage was way too high.
Interesting. Maybe something to do with bond convexity (as rates get lower they get more sensitive); although that should have the opposite effect as to what you are suggesting.

dspencer
Posts: 195
Joined: Wed Jul 06, 2016 11:29 am

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

Post by dspencer » Fri Sep 13, 2019 12:13 pm

MotoTrojan wrote:
Fri Sep 13, 2019 8:46 am
dspencer wrote:
Fri Sep 13, 2019 8:36 am

MotoTrojan,

Are you essentially considering EDV as a leveraged fund even though it is technically not leveraged? What are you considering it's effective leverage to be?
Essentially yes, although the 22% calc is simply the reduction in UPRO exposure. On the equity side you are leveraging to boost returns, on the bond side what you are really doing is leveraging to boost the effective-duration. Looking purely at how much leverage as a % you are using on the bond side is somewhat flawed as 3x leverage on 20 year bonds is much different than 3x leverage on short-term treasuries. If we use 55/45 UPRO/TMF as the comparison point, 43/57 UPRO/EDV is a 22% reduction in equity exposure/leverage, with an effective-duration ratio held constant (again, I used volatility via Simba from 1955-2018 as a proxy for duration).

Regardless of how I got there, I like how this allocation behaves through back-testing via Simba as well as the data since VEDTX inception (2008 I believe). It gives me treasury exposure rather than corporates (such as Pimco StocksPLUS) and it edges up the equity exposure somewhat, which will boost the overall expected return while reducing the exposure the rates. Most importantly compared to the OPs strategy though is that it eliminates the path dependency on the bond side; if 20 year rates stay where they are today for the next 30 years, a portfolio holding unleveraged S&P500/LTT may still have a great Sharpe ratio, but the portfolio using TMF could get crushed due to volatility decay, where-as the EDV holding will providing some income and rebalancing bonus without said decay. Even if rates work their way down to 0% over 30 years, EDV could still very likely come out ahead with enough bond volatility.

If we (IMHO wrongly) assume that 55/45 UPRO/TMF will perform similarly moving forward as it has in the past 37 years (20% CAGR) then I am still going to have monster performance (17% CAGR), and even if things are closer to 1955-present I should still beat the S&P500 by 1-2% (while OP may be more like 2-3%), but there are several scenarios where I beat the S&P500 while the OP doesn't.
Thanks for the explanation. I plan to shift some of my TMF allocation over to EDV. My rationale is:

1. I'd like to slightly increase my equity ratio, although probably not to 55%. At 40/60 with rates this low it feels like the portfolio is almost more of a bear bet than a bull bet.
2. I'd like to reduce my overall leverage. I'm not sure my confidence justifies 3x leverage and also not sure that 3x is optimal.

I'm considering 40/30/30 (UPRO/TMF/EDV). If any of the smarter folks have an opinion I'm happy to hear it.

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Fri Sep 13, 2019 12:15 pm

dspencer wrote:
Fri Sep 13, 2019 12:13 pm
MotoTrojan wrote:
Fri Sep 13, 2019 8:46 am
dspencer wrote:
Fri Sep 13, 2019 8:36 am

MotoTrojan,

Are you essentially considering EDV as a leveraged fund even though it is technically not leveraged? What are you considering it's effective leverage to be?
Essentially yes, although the 22% calc is simply the reduction in UPRO exposure. On the equity side you are leveraging to boost returns, on the bond side what you are really doing is leveraging to boost the effective-duration. Looking purely at how much leverage as a % you are using on the bond side is somewhat flawed as 3x leverage on 20 year bonds is much different than 3x leverage on short-term treasuries. If we use 55/45 UPRO/TMF as the comparison point, 43/57 UPRO/EDV is a 22% reduction in equity exposure/leverage, with an effective-duration ratio held constant (again, I used volatility via Simba from 1955-2018 as a proxy for duration).

Regardless of how I got there, I like how this allocation behaves through back-testing via Simba as well as the data since VEDTX inception (2008 I believe). It gives me treasury exposure rather than corporates (such as Pimco StocksPLUS) and it edges up the equity exposure somewhat, which will boost the overall expected return while reducing the exposure the rates. Most importantly compared to the OPs strategy though is that it eliminates the path dependency on the bond side; if 20 year rates stay where they are today for the next 30 years, a portfolio holding unleveraged S&P500/LTT may still have a great Sharpe ratio, but the portfolio using TMF could get crushed due to volatility decay, where-as the EDV holding will providing some income and rebalancing bonus without said decay. Even if rates work their way down to 0% over 30 years, EDV could still very likely come out ahead with enough bond volatility.

If we (IMHO wrongly) assume that 55/45 UPRO/TMF will perform similarly moving forward as it has in the past 37 years (20% CAGR) then I am still going to have monster performance (17% CAGR), and even if things are closer to 1955-present I should still beat the S&P500 by 1-2% (while OP may be more like 2-3%), but there are several scenarios where I beat the S&P500 while the OP doesn't.
Thanks for the explanation. I plan to shift some of my TMF allocation over to EDV. My rationale is:

1. I'd like to slightly increase my equity ratio, although probably not to 55%. At 40/60 with rates this low it feels like the portfolio is almost more of a bear bet than a bull bet.
2. I'd like to reduce my overall leverage. I'm not sure my confidence justifies 3x leverage and also not sure that 3x is optimal.

I'm considering 40/30/30 (UPRO/TMF/EDV). If any of the smarter folks have an opinion I'm happy to hear it.
I also considered that exact portfolio for some time. Have you downloaded the Simba spreadsheet to play around with these allocations? Or even just VEDTX (captures most of the GFC) and simulated UPRO/TMF data? 40/60 UPRO/EDV would have similar drawdown to 100% S&P500 in the GFC, so unless you want to capture the cap-gains of rates going down further, I'd just fully adopt the efficiency of EDV.

EfficientInvestor
Posts: 202
Joined: Thu Nov 01, 2018 7:02 pm

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

Post by EfficientInvestor » Fri Sep 13, 2019 12:59 pm

MotoTrojan wrote:
Fri Sep 13, 2019 11:44 am
rascott wrote:
Fri Sep 13, 2019 11:31 am
MotoTrojan wrote:
Fri Sep 13, 2019 11:27 am
rascott wrote:
Fri Sep 13, 2019 11:25 am
I've decided to eventually transition my lottery ticket play to the futures market. Eliminates the mgmt fee, also eliminates the volatility decay of TMF.

150% SP500
1500% Short-Term Treasuries

This isn't a static leverage level...in higher rate environments one would dial down the leverage of STTs to meet the target duration.

From 2010 (current rate world)

https://www.portfoliovisualizer.com/bac ... total3=100

From early 90s....using the long-run ratio of 6:1 STTs/Equities.

https://www.portfoliovisualizer.com/bac ... total3=100
Why does duration change based on rate? That doesn't make sense.
Probably not the right terminology. But when rates are lower, you have to leverage STTs at a higher rate for them to equalize the volatility of LTTs. Maybe someone smarter than I can explain that.

SHY has a duration of 1.89 currently. If I wanted to equate that to TMF, I'd need to lever it 28X. 60% (original risk parity TMF allocation) of 28x leverage gets me at roughly 16.8....or 1680% STTs.

But in eras of higher rates, this level of leverage was way too high.
Interesting. Maybe something to do with bond convexity (as rates get lower they get more sensitive); although that should have the opposite effect as to what you are suggesting.
You are correct in saying that convexity would have the opposite effect and causes bonds to be more volatile at lower interest rates. However, convexity does not have much effect on short term treasuries. Notice how the shorter term maturity lines (purple lines) in the graph below have a constant separation and don't curve away from each other as interest rates decrease. In other words, the volatility of STTs is the same regardless of interest rates. (Thanks Tyler9000 for the great chart!)

Image
Source: https://portfoliocharts.com/2019/05/27/ ... convexity/

However, once you start borrowing money to invest in STT, your volatility becomes dependent upon how much you borrow and the rate at which you borrow. I put together the graph below to get an idea of how the risk-free rate affects volatility of leveraged STTs. As risk-free rate increases, so does volatility (all the lines have a positive slope). As you increase leverage, the volatility increases at a faster rate (the slopes are steeper). Notice that the stock market has a flat line around 15% volatility and there isn't much correlation between risk-free rate and volatility for stock.

This chart shows that in order to attempt to achieve risk parity between stocks and leveraged STT, you should change the amount of leverage used on STT in an attempt to target the volatility of stocks. For instance, since we are currently around a 2% risk free rate, you may want to use 10x leverage on the STT because the 10x line (orange) intersects the S&P volatility line (black) at 2%.

Addition: I will also add that I made similar charts for intermediate and long term treasuries. The lines in those charts only had slightly positive slopes. My assumption is that for longer maturities, the effect of convexity counteracts the effect of leverage and you wind up with a more constant volatility regardless of risk-free rate.

Image
Last edited by EfficientInvestor on Fri Sep 13, 2019 2:23 pm, edited 2 times in total.

rascott
Posts: 702
Joined: Wed Apr 15, 2015 10:53 am

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

Post by rascott » Fri Sep 13, 2019 1:24 pm

EfficientInvestor wrote:
Fri Sep 13, 2019 12:59 pm
MotoTrojan wrote:
Fri Sep 13, 2019 11:44 am
rascott wrote:
Fri Sep 13, 2019 11:31 am
MotoTrojan wrote:
Fri Sep 13, 2019 11:27 am
rascott wrote:
Fri Sep 13, 2019 11:25 am
I've decided to eventually transition my lottery ticket play to the futures market. Eliminates the mgmt fee, also eliminates the volatility decay of TMF.

150% SP500
1500% Short-Term Treasuries

This isn't a static leverage level...in higher rate environments one would dial down the leverage of STTs to meet the target duration.

From 2010 (current rate world)

https://www.portfoliovisualizer.com/bac ... total3=100

From early 90s....using the long-run ratio of 6:1 STTs/Equities.

https://www.portfoliovisualizer.com/bac ... total3=100
Why does duration change based on rate? That doesn't make sense.
Probably not the right terminology. But when rates are lower, you have to leverage STTs at a higher rate for them to equalize the volatility of LTTs. Maybe someone smarter than I can explain that.

SHY has a duration of 1.89 currently. If I wanted to equate that to TMF, I'd need to lever it 28X. 60% (original risk parity TMF allocation) of 28x leverage gets me at roughly 16.8....or 1680% STTs.

But in eras of higher rates, this level of leverage was way too high.
Interesting. Maybe something to do with bond convexity (as rates get lower they get more sensitive); although that should have the opposite effect as to what you are suggesting.
You are correct in saying that convexity would have the opposite effect and causes bonds to be more volatile at lower interest rates. However, convexity does not have much effect on short term treasuries. Notice how the shorter term maturity lines (purple lines) in the graph below have a constant separation and don't curve away from each other as interest rates decrease. In other words, the volatility of STTs is the same regardless of interest rates. (Thanks Tyler9000 for the great chart!)

Image
Source: https://portfoliocharts.com/2019/05/27/ ... convexity/

However, once you start borrowing money to invest in STT, your volatility becomes dependent upon how much you borrow and the rate at which you borrow. I put together the graph below to get an idea of how the risk-free rate affects volatility of leveraged STTs. As risk-free rate increases, so does volatility (all the lines have a positive slope). As you increase leverage, the volatility increases at a faster rate (the slopes are steeper). Notice that the stock market has a flat line around 15% volatility and there isn't much correlation between risk-free rate and volatility for stock.

This chart shows that in order to attempt to achieve risk parity between stocks and leveraged STT, you should change the amount of leverage used on STT in an attempt to target the volatility of stocks. For instance, since we are currently around a 2% risk free rate, you may want to use 10x leverage on the STT because the 10x line (orange) intersects the S&P volatility line (black) at 2%.

Image

Great graph! Thank you! May need to print that out and pin to my office wall!

I got to a 150/1500 allocation as well for risk parity, but in a much more round- about way than just looking at that chart.

Lee_WSP
Posts: 816
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 Sep 13, 2019 2:35 pm

Why EDV and UPRO instead of EDV and SSO?

caklim00
Posts: 1840
Joined: Mon May 26, 2008 10:09 am

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

Post by caklim00 » Fri Sep 13, 2019 3:55 pm

Almost back to no gain after the beatdown that TMF has taken recently.

My current rankings:
1) EDV/UPRO: 57/43
2) TMF/UPRO: 45/55
3) 20 day lookback for risk parity
4) 20% Volatility 20 day lookback

But, SLYV and ISCF which I own a ton of are finally doing well :)

Lee_WSP
Posts: 816
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 Sep 13, 2019 5:41 pm

This week has really brought home my reservations about the "spikiness" of TMF. EDV is looking a lot less "prickly".

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Fri Sep 13, 2019 6:00 pm

Lee_WSP wrote:
Fri Sep 13, 2019 2:35 pm
Why EDV and UPRO instead of EDV and SSO?
More leverage. If I used SSO I couldn’t hold as much EDV for the same equity exposure.

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Fri Sep 13, 2019 6:01 pm

caklim00 wrote:
Fri Sep 13, 2019 3:55 pm
Almost back to no gain after the beatdown that TMF has taken recently.

My current rankings:
1) EDV/UPRO: 57/43
2) TMF/UPRO: 45/55
3) 20 day lookback for risk parity
4) 20% Volatility 20 day lookback

But, SLYV and ISCF which I own a ton of are finally doing well :)
You’re welcome ;).

eleventhstreet
Posts: 15
Joined: Tue Jul 23, 2019 10:54 pm

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

Post by eleventhstreet » Fri Sep 13, 2019 7:09 pm

I really liked everyone's question about "what if Japan".
It is an interesting scenario we have not yet experienced in the US: What if there was no crash but there was just very stagnant growth and minimal interest rates. In that case this kind of a portfolio would be burned by the interest costs. (In other words if the SP500 just kinda cruised, interest rates kinda just cruised, and all you got was a margin bill to pay)

Just curious has anyone tried back testing this portfolio through various different countries economies? It would be fun to note because if you think about it, the US is just one country and I can only think of 5 or 6 unique scenarios our economy has gone through. But the global economy has hundreds of countries with unique economies that end up in funny unique scenarios (e.g. hyperinflation, stagnation, etc etc)

MotoTrojan
Posts: 5865
Joined: Wed Feb 01, 2017 8:39 pm

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

Post by MotoTrojan » Fri Sep 13, 2019 7:11 pm

eleventhstreet wrote:
Fri Sep 13, 2019 7:09 pm
I really liked everyone's question about "what if Japan".
It is an interesting scenario we have not yet experienced in the US: What if there was no crash but there was just very stagnant growth and minimal interest rates. In that case this kind of a portfolio would be burned by the interest costs. (In other words if the SP500 just kinda cruised, interest rates kinda just cruised, and all you got was a margin bill to pay)

Don't forget the volatility decay!

User avatar
queso
Posts: 736
Joined: Thu Jan 07, 2016 3:52 pm

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

Post by queso » Fri Sep 13, 2019 7:12 pm

Lee_WSP wrote:
Fri Sep 13, 2019 5:41 pm
This week has really brought home my reservations about the "spikiness" of TMF. EDV is looking a lot less "prickly".
+1. My hedgefundie fundie ain't doing so hot compared to VTSAX. *sigh*

After years of disappointments with get rich quick schemes I know I'm gonna get rich with this scheme, and quick! :happy

Topic Author
HEDGEFUNDIE
Posts: 3286
Joined: Sun Oct 22, 2017 2:06 pm

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

Post by HEDGEFUNDIE » Fri Sep 13, 2019 7:14 pm

queso wrote:
Fri Sep 13, 2019 7:12 pm
Lee_WSP wrote:
Fri Sep 13, 2019 5:41 pm
This week has really brought home my reservations about the "spikiness" of TMF. EDV is looking a lot less "prickly".
+1. My hedgefundie fundie ain't doing so hot compared to VTSAX. *sigh*

After years of disappointments with get rich quick schemes I know I'm gonna get rich with this scheme, and quick! :happy
Stay. The. Course.

Post Reply