HEDGEFUNDIE's excellent adventure Part II: The next journey

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Thereum
Posts: 60
Joined: Sun Jun 14, 2020 9:05 pm

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

Post by Thereum »

What's the argument for using leveraged ETFs instead of using non-leveraged ETFs on margin? I've done some backtesting and found better returns using margin. Is the goal just to avoid the possibility of a margin call? I suppose that is a good reason for most investors.
RayKeynes
Posts: 259
Joined: Mon Nov 11, 2019 2:14 am

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

Post by RayKeynes »

Dear guys,

I think the idea is very interesting, some questions:

1. What is the return YTD for this strategy compared to the SP500 total return?
2. What do you guys think of implementing the following strategy:

a) If SP500 trades above SMA220-235, buy 100% UPRO
b) If SP500 trades below SMA220-235, sell 50% of the portfolio value and hold cash. With the other 50%, buy 60% TMF and 40% UPRO.

The fact of using 100% UPRO above the SMA will increase the return going forward drastically.
SVT
Posts: 359
Joined: Mon Oct 13, 2008 8:56 am

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

Post by SVT »

RayKeynes wrote: Wed Jun 17, 2020 1:41 am Dear guys,

I think the idea is very interesting, some questions:

1. What is the return YTD for this strategy compared to the SP500 total return?
I'm not sure about YTD, as it doesn't appear that M1 has that option but since I started in Feb 2019, I'm up exactly 100%, the past 1 yr up 67%, the last 3 months up 42%. You can look up what the S&P has done over those time frames.

The return of this is going to be different for different people depending on their exact asset allocation to these 2 funds and the timing of rebalancing. I stuck with the 40/60 much later than the OP, up till I think just a couple of months ago and then went 50/50. I've since drifted to 31/69 TMF/UPRO. I will rebalance on or about 7/1.
RayKeynes
Posts: 259
Joined: Mon Nov 11, 2019 2:14 am

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

Post by RayKeynes »

SVT wrote: Wed Jun 17, 2020 5:39 am
RayKeynes wrote: Wed Jun 17, 2020 1:41 am Dear guys,

I think the idea is very interesting, some questions:

1. What is the return YTD for this strategy compared to the SP500 total return?
I'm not sure about YTD, as it doesn't appear that M1 has that option but since I started in Feb 2019, I'm up exactly 100%, the past 1 yr up 67%, the last 3 months up 42%. You can look up what the S&P has done over those time frames.

The return of this is going to be different for different people depending on their exact asset allocation to these 2 funds and the timing of rebalancing. I stuck with the 40/60 much later than the OP, up till I think just a couple of months ago and then went 50/50. I've since drifted to 31/69 TMF/UPRO. I will rebalance on or about 7/1.
Really incredible performance! Congratulations.

I am, to be honest, more on the side of UPROSMA strategy, which works as follows:
- BUY and HOLD UPRO if GSPC above SMA230
- SELL and HOLD CASH if GSPC below SMA230

This strategy has outperformed UPRO / TMF from 2010 - 2019 by significant amounts. However, due to the beginning of the corona crysis in February 2020, one could achieve a 100% return on UPRO/TMF while only a 0% return on UPROSMA
hillman
Posts: 189
Joined: Thu Jun 07, 2012 2:08 pm

HEDGEFUNDIE's excellent adventure at Marketwatch.com

Post by hillman »

[Thread merged into here, see below. --admin LadyGeek]

I just saw this post and thought it was funny that UPRO and TMF are named. However, they are discussing a 50/50 mix. Obviously, they have not read the multiple posts and millions of comments refining the plan.

https://www.marketwatch.com/story/how-3 ... yptr=yahoo
Semantics
Posts: 168
Joined: Tue Mar 10, 2020 1:42 pm

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

Post by Semantics »

Thereum wrote: Sun Jun 14, 2020 9:26 pm What do you guys think about buying SWAN on margin? SWAN is roughly 70% equities, 90% 10-year treasuries, but because it uses call options, it is long volatility. Volatility tends to expand during downturns, which offsets losses due to the positive delta of the call option. The call option is short theta, but because it has a long expiry and is in the money, the actual theta decay is very low. A big advantage of leveraging SWAN instead of UPRO/TMF is that you won't suffer from the daily compounding decay (which I will concede helps as often as it hurts).

Anyway, if I compare 55/45 UPRO/TMF to 235% SWAN, I get better results in terms of Sharpe ratio. (I leveraged SWAN by this amount to equal the equity exposure.) The Sharpe ratio implies that SWAN could be leveraged much further without blowing up. I also did some tests with the historical SWAN index data going back to 2005 and found that it would easily survive 15x leverage.
I found this idea enticing, I like the idea of bounding my downside, but looking into it a bit more I don't think the returns would be amplified enough for it to be worth it.

- SWAN only has 1.5 years of data, during which we had the ideal conditions for it to outperform, so I am wary of overfitting using just that time period to backtest, thus it's helpful to look at the SWANXT index.
- CASHX doesn't factor in borrowing costs, which if returns drop from 2010s levels and/or the risk-free rate goes up are going to be a drag on the portfolio. The leveraged ETFs have this problem too, but just something to keep in mind when using PV. I recently saw there's a whole thread on modeling leveraged ETFs more accurately.
- Spreadsheeting the SWANXT data, returns since 2006 have roughly matched returns of S&P 500 *before* expense ratio (with much lower volatility of course). During that time though, the average risk-free rate was something like 2.5%, which would cut the returns on a leveraged part a lot. With 2.35x leverage reset monthly I get 13.46% CAGR, which turns into 8.91% after subtracting 0.5*2.35 (ER) and 2.5*1.35 (risk free rate approx), vs 8.5% for the S&P 500. Std dev is still lower (~12% vs ~15%), but the Sharpe ratio is almost certainly worse than UPRO/TMF, and this isn't really much different than an 80/20 portfolio.

This is all napkin math which is a big caveat, but if I'm not horribly off, I can think of a few reasons this ends up looking uncompelling:

- SWAN's equity leverage is not always 7x, maybe that's typical, but right now the Dec 2021 0.7 delta call option gives 4x leverage (probably should have looked at June 2021 here but I think it'd be similar).
- With leveraged SWAN you're paying borrowing costs to pay borrowing costs (implicit in the options), which is a bit of added inefficiency.
- In a bull market the leverage decreases over time as the options go further ITM, lowering the relative returns of the stock component.
- Intuitively, I suppose there's only so much reward you can squeeze out of any given level of risk with mixing/matching the same assets. The UPRO/TMF portfolio amps up the volatility more, and the returns follow suit. Leveraging SWAN something like 6-8x would match the volatility but still return less because of the previous point, unless the market stays flat/declines.
ChrisBenn
Posts: 453
Joined: Mon Aug 05, 2019 7:56 pm

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

Post by ChrisBenn »

Thereum wrote: Sun Jun 14, 2020 9:26 pm What do you guys think about buying SWAN on margin? SWAN is roughly 70% equities, 90% 10-year treasuries, but because it uses call options, it is long volatility. Volatility tends to expand during downturns, which offsets losses due to the positive delta of the call option. The call option is short theta, but because it has a long expiry and is in the money, the actual theta decay is very low. A big advantage of leveraging SWAN instead of UPRO/TMF is that you won't suffer from the daily compounding decay (which I will concede helps as often as it hurts).

(...)
I would take a step back and consider what you are implicitly doing when you do that. Loot at swan at time of rebalance, where it's 90% treasuries and 10% call options at a target .7 delta.

When you buy on margin the majority of what you are purchasing is treasuries. I'm going to guess with a pretty high confidence that your margin rates are higher than the treasury return.

I think you might be better served to figure out the effective equity exposure you would get through the 10% call options (in swan) X your target leverage, and just buy those options directly. Since you can't get 10x leverage on margin you won't be spending all of your investment on options either way. The rest you then would then place in something else.

If you want the same volatility dampening from assumed negative correlation it's the same discussion as the EA here. GOVT gives you broad duration treasury exposure in one fund, but given you are spending more than 10% of your investment on call options, you would not have the same effective treasury:equity ratio. EDV is the best you can do unleveraged. TMF is the best you can do in the ETF leveraged space. If you want to avoid the daily resets, and EDV isn't sufficient then I think treasury futures are your only option. Big issue there is contract size; if you aren't in the mid 6 figures at least for investment it makes re-balancing annoying

Also see the SWAN thread for some points brought up at the end about expectations. If the markets are flat you are going to bleed theta on those options. It was pointed out that "flat" might be a misnomer, and you might need larger gains than you expect just to break even. If you are rolling the options yourself at the retail level you can maybe do better with 2 year leaps and selling at 1 year so you don't burn as much theta, but still keep ltcg.

(This is another point - while you can keep ltcg, you are still forced to realize any gains now, instead of deferring indefinitely - since you have to roll the leaps)

I also think it might be difficult to get the 200%+ margin you are looking at, unless you are going portfolio margin. At that point you are def. risking a bit more.

Also, re: swan on margin note the anomalous spike around the 20th or march
Image
https://stockcharts.com/h-perf/ui?s=SWA ... 4133020142

It corrected very quickly the next day (seemed to be some weird bid up before closing), but if that spike was in the other direction Interactive Brokers, for instance, would have probably auto-liquidated part of your holdings to cover it if you were highly margined up.
orvanik
Posts: 11
Joined: Thu Apr 09, 2020 9:49 am

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

Post by orvanik »

To all you VIX hopeful hedgers I have finally read the paper that was posted a while back and I am back with my promised summary.

Copy of the paper can be found here: https://papers.ssrn.com/sol3/papers.cfm ... id=3514141

Warning: I know very little, this is not investment advice, I hope I did not royally misconstrue anything said in the paper or by anyone else I reference.

Summary: If the VIX were an investible asset, when combined with a S&P 500 ETF, can provide a strong hedge, significant reduction in volatility, and an overall increase in long-term CAGR.

Notes: All statistics come from the paper unless otherwise noted and the sampling period is from 1990 ending in 2017.
· Correlation is -0.64
· In months when the S&P return was positive, the average return for S&P and VIX was 3.07% and -6.83% respectively.
· When S&P was negative, returns for S&P and VIX were -3.45% and 16.25% respectively.
· 75% of months, the S&P and VIX have different signed returns. 46% of those the S&P is up while the VIX is down, 17% of the sample has months when both the S&P and VIX are positively signed. 8% of months in the sample both the S&P and VIX were negatively signed. The assets are not always negatively correlated.
· ~ for every 1% decrease in the S&P, the VIX will increase by 5.36%, but for every 1% gain the VIX may not move at all. It is the VIX's asymmetric movement and mean reversion that makes it a good hedge.
· The paper targets a VIX allocation that would eliminate a 5% drawdown in the S&P. Because of the VIX's asymmetry the necessary allocation will positively correlate with the level of the VIX.
· The VIX showed a 33% return at VIX of 15 when the S&P returned -5%, but at a VIX of 28, the move is more muted as a 5% move is more likely the higher the VIX.
· The allocation to the VIX required to eliminate a 5% drawdown in the S&P therefore varied from 13% to 30%, with an average of ~20%.
· This makes sense based on the research done by Chris Cole at Artemis Capital. In his most recent paper, "The Allegory of the Hawk and Serpent," he suggests a ~21% allocation to long volatility, which is the trade being made here by creating a proxy to go long the VIX. His paper can be found here: https://www.artemiscm.com/welcome#research
· The researchers tested the optimal allocation to the VIX for every period in the sample, using a 60 month regression, where in the case of positive S&P returns the optional allocation would be 0. The average allocation came out to be 5.05%
· This is a significantly lower allocation than suggested by Chris Cole, however, the portfolio here being tested is strictly a combination of the S&P 500 and VIX, vs his Dragon Portfolio which is significantly different.
· The 5% average allocation here can be more aptly compared to NNT's and Mark Spitznagel's Universa suggested allocation of 3.33% of assets to their tail risk fund. They are likely to achieve better convexity and therefore require a lower allocation to achieve the same level of results.
· Due to the mean reverting nature of the VIX, above a certain threshold it makes sense to exit the hedge, or significantly reduce exposure. This also means that the cost of holding the hedge when the VIX is at below average levels even during a rising market is not likely to cause a significant drag on performance.
· Using no threshold, a portfolio that holds ~5% VIX shows alpha of almost 1% per year, however the result is insignificant. The alpha is significant after controlling for certain factors which the researchers tested including CAPM, Fama-French, Momentum, and two option based factors, however, I am less interested in this result. For this to be effective for an individual, I would want to see significance before controlling for CAPM and FF.
· This makes choosing threshold level to remove the hedge very important. The researches choose three threshold levels to test, VIX[MEAN], VIX[1-SD], and VIX[1+SD]. As would be expect VIX[1-SD] is too low and the hedge is not effective. However, too high a threshold creates the same effect as having no threshold and the mean reversion hurts returns.
· VIX[1+SD] was most effective and reduced CAPM beta to .81. They also tested this in two sub periods, 2007-2011, and 2013-2017. The first with high volatility and the second with very low sustained volatility. Both periods show significant alphas which shows that holding the VIX even during periods of sustained low volatility can have benefits.
· Now the big problem that has been documented before, the VIX index is not directly investable and the volatility instruments that investors seek as alternatives like VXX, VIX futures and others end up underperforming the S&P 500.
· The researchers found that the cheapest and most effective way to replicate the VIX was to buy a monthly ITM-OTM put spread and sell an ATM-OTM call spread. However this has the downside of underperforming during large upswings. I also have not been able to figure out exactly what spreads the researchers simulated and therefore do not yet know how to implement this personally. It appears based on the exhibits that the researchers looked at ITM and OTM options at strikes 2.5% and 5% from spot prices.
·"This call spread not only helps to fund the volatility and price protection, but when combined with the net long equity position will have a payoff structure similar to a covered call."
· They bought these options at the beginning of month t that expires in month t+1 and closed them at the end of month t.
· Using the threshold of VIX[1+SD] over the sample period of 1997-2017, the portfolio returned 10.1% pa with volatility of 15.2%. The return is 1.8% higher than the S&P while being 3.7% less risky. It underperforms the cash VIX portfolio by 1.3% which is expected.
· The excess returns come from outperformance in negative markets as would be expected. On average there was 18% outperformance in negative markets and only a 5% drag in positive markets. In 2000 and 2008, the portfolio outperformed by 35% and 34% respectively.
· This makes sense as losses are significantly more important in growing capital. The first rule of investing is don't lose money.
· Additionally, the put portfolio actually outperformed a theoretical cash VIX portfolio in some high volatility markets as the puts maintain their value even after volatility falls.
· However, since 2011, the portfolio has underperformed the S&P on a return, but not risk adjusted basis. This is due to the underperformance of covered calls. Note the sample period ends in 2017, so this does not include the end of 2018 or the most recent VIX spike in Feb/Mar which would have triggered the threshold.
· The VIX threshold level does not meaningfully matter as long as it is somewhere between the mean level and mean plus two SD.
User avatar
LadyGeek
Site Admin
Posts: 67337
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: HEDGEFUNDIE's excellent adventure at Marketwatch.com

Post by LadyGeek »

I merged hillman's post into the on-going discussion (marketwatch article already posted in this thread by randyharris).

^^^ Hannah Rodriguez's post is in response to:
hillman wrote: Wed Jun 17, 2020 12:12 pm [Thread merged into here, see below. --admin LadyGeek]

I just saw this post and thought it was funny that UPRO and TMF are named. However, they are discussing a 50/50 mix. Obviously, they have not read the multiple posts and millions of comments refining the plan.

https://www.marketwatch.com/story/how-3 ... yptr=yahoo
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
Zb3
Posts: 63
Joined: Thu Sep 25, 2014 11:25 pm

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

Post by Zb3 »

Anyone notice that the s&p 500 is 10 percent off it’s all time high but UPRO is 80 percent off its high. All down to volatility decay I guess ?
Lee_WSP
Posts: 3465
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

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

Post by Lee_WSP »

Zb3 wrote: Fri Jun 19, 2020 5:03 pm Anyone notice that the s&p 500 is 10 percent off it’s all time high but UPRO is 80 percent off its high. All down to volatility decay I guess ?
Yes. Yes, but the high was pretty lofty to begin with.
LittleBitMore
Posts: 89
Joined: Mon Dec 23, 2019 3:44 pm

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

Post by LittleBitMore »

Zb3 wrote: Fri Jun 19, 2020 5:03 pm Anyone notice that the s&p 500 is 10 percent off it’s all time high but UPRO is 80 percent off its high. All down to volatility decay I guess ?
Not really the right phrasing -- its 45% off its high, 80% increase away from hitting it again. Otherwise when it was at 17 it would have been "300% off its all time high"
User avatar
coingaroo
Posts: 93
Joined: Fri Apr 26, 2019 11:31 am

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

Post by coingaroo »

+1 to ChrisBenn. If you're going to buy SWAN on margin, might as well as just buy SPY options yourself. A huge risk of buying SWAN on margin is that market makers might bite the dust one day (for one reason or another), ETF goes way down, and you get liquidated. The "reward" of buying SWAN is you... save yourself 5 minutes of work every month. The risk is continued management fee bleed, counterparty risk, and portfolio liquidation risk. No thanks!

--

I would like to ask if there is any convenient & affordable way to backtest a TQQQ + rolling VIX OTM call strategy.

My impetus is pretty simple: while TQQQ/TMF has worked in the past, I do believe a run-up on inflation is not out of the question. TQQQ/VIX calls seem like they have the potential to be a more reliable option.
langlands
Posts: 612
Joined: Wed Apr 03, 2019 10:05 pm

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

Post by langlands »

For people who are considering adding VIX to your portfolio, is there a reason you aren't considering simply adding out of the money put options on SPX? Several posts are talking about VIX calls, which seems way more complicated than an option directly on the SPX.
User avatar
physixfan
Posts: 210
Joined: Wed Apr 03, 2019 4:09 pm

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

Post by physixfan »

coingaroo wrote: Sat Jun 20, 2020 8:32 am +1 to ChrisBenn. If you're going to buy SWAN on margin, might as well as just buy SPY options yourself. A huge risk of buying SWAN on margin is that market makers might bite the dust one day (for one reason or another), ETF goes way down, and you get liquidated. The "reward" of buying SWAN is you... save yourself 5 minutes of work every month. The risk is continued management fee bleed, counterparty risk, and portfolio liquidation risk. No thanks!

--

I would like to ask if there is any convenient & affordable way to backtest a TQQQ + rolling VIX OTM call strategy.

My impetus is pretty simple: while TQQQ/TMF has worked in the past, I do believe a run-up on inflation is not out of the question. TQQQ/VIX calls seem like they have the potential to be a more reliable option.
Since all the VIX ETNs consistently lose value over time significantly, I think it is quite certain that the contango is huge even if you roll VIX calls by yourself... That's way worse than losing money on management fees on ETFs like SWAN in my opinion...
genexsamples
Posts: 1
Joined: Sat Jun 20, 2020 12:46 pm

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

Post by genexsamples »

Hi all, I've been following this thread for a while now, but finally posting!

I'm interested in building a portfolio that takes advantage of holding and rebalancing out of VIX, much like what is described Doran 2020.

I have access to historical option price data going back to 1996, and I'm first going to implement the model they describe in the paper, then start tweaking it to my preferences. If anyone has "hot takes" on the Doran paper, or wants to help figuring out the best way to implement a strategy like theirs, please chime in.

Thanks & good luck!
User avatar
coingaroo
Posts: 93
Joined: Fri Apr 26, 2019 11:31 am

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

Post by coingaroo »

genexsamples wrote: Sat Jun 20, 2020 1:34 pm Hi all, I've been following this thread for a while now, but finally posting!

I'm interested in building a portfolio that takes advantage of holding and rebalancing out of VIX, much like what is described Doran 2020.

I have access to historical option price data going back to 1996, and I'm first going to implement the model they describe in the paper, then start tweaking it to my preferences. If anyone has "hot takes" on the Doran paper, or wants to help figuring out the best way to implement a strategy like theirs, please chime in.

Thanks & good luck!
May I ask how you got access to the option data? I’m interested in exploring it as well.
Rad Dad
Posts: 30
Joined: Fri Aug 09, 2019 9:48 pm

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

Post by Rad Dad »

langlands wrote: Sat Jun 20, 2020 11:59 am For people who are considering adding VIX to your portfolio, is there a reason you aren't considering simply adding out of the money put options on SPX? Several posts are talking about VIX calls, which seems way more complicated than an option directly on the SPX.
I initially assumed that implementing VIX calls was going to be really complicated, but after reading this article it doesn’t sound too complicated at all.

http://www.cboe.com/blogs/options-hub/2 ... vxth-index

I’m interested in finding a way to replace a small portion of my TMF allocation with an alternative hedge so I’m more than willing to hear why one is better than the other.
Semantics
Posts: 168
Joined: Tue Mar 10, 2020 1:42 pm

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

Post by Semantics »

physixfan wrote: Sat Jun 20, 2020 12:21 pm
coingaroo wrote: Sat Jun 20, 2020 8:32 am +1 to ChrisBenn. If you're going to buy SWAN on margin, might as well as just buy SPY options yourself. A huge risk of buying SWAN on margin is that market makers might bite the dust one day (for one reason or another), ETF goes way down, and you get liquidated. The "reward" of buying SWAN is you... save yourself 5 minutes of work every month. The risk is continued management fee bleed, counterparty risk, and portfolio liquidation risk. No thanks!

--

I would like to ask if there is any convenient & affordable way to backtest a TQQQ + rolling VIX OTM call strategy.

My impetus is pretty simple: while TQQQ/TMF has worked in the past, I do believe a run-up on inflation is not out of the question. TQQQ/VIX calls seem like they have the potential to be a more reliable option.
Since all the VIX ETNs consistently lose value over time significantly, I think it is quite certain that the contango is huge even if you roll VIX calls by yourself... That's way worse than losing money on management fees on ETFs like SWAN in my opinion...
Note that VIX futures are in slight backwardation right now, so I would assume the (unleveraged) funds that track it might actually be gaining a bit over time. During less volatile periods they do lose value to contango like you say, presumably slightly less now than pre-pandemic since rates are lower.
langlands
Posts: 612
Joined: Wed Apr 03, 2019 10:05 pm

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

Post by langlands »

Rad Dad wrote: Sat Jun 20, 2020 2:56 pm
langlands wrote: Sat Jun 20, 2020 11:59 am For people who are considering adding VIX to your portfolio, is there a reason you aren't considering simply adding out of the money put options on SPX? Several posts are talking about VIX calls, which seems way more complicated than an option directly on the SPX.
I initially assumed that implementing VIX calls was going to be really complicated, but after reading this article it doesn’t sound too complicated at all.

http://www.cboe.com/blogs/options-hub/2 ... vxth-index

I’m interested in finding a way to replace a small portion of my TMF allocation with an alternative hedge so I’m more than willing to hear why one is better than the other.
In terms of the literal mechanics of implementing the trade, adding VIX calls into your portfolio probably isn't much more complicated than adding SPX puts into your portfolio considering both securities are readily available at most brokerages. I'm more concerned about the fact that VIX calls are a "2nd order" instrument whereas SPX puts are only a "1st order" instrument (and the SPX index would be a "0th order" instrument). What I mean by that is that to use SPX puts, you need to understand the Black-Scholes formula (which is only itself an approximation) and how volatility is priced, and what a volatility surface is etc. The theory is quite complex mathematically. To understand VIX calls, you need to understand how vol of vol is priced. This brings in even more complexity, and as someone who did an intern project once on stochastic volatility models, this stuff isn't easy and a lot of it is still highly academic.

When I'm constructing my portfolio, thinking about hedging with volatility is complex enough. I'm just skeptical about adding in volatility of volatility itself as an additional complexity to consider.
HedgeFundMillionaire
Posts: 12
Joined: Tue Mar 24, 2020 11:30 am

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

Post by HedgeFundMillionaire »

Zb3 wrote: Fri Jun 19, 2020 5:03 pm Anyone notice that the s&p 500 is 10 percent off it’s all time high but UPRO is 80 percent off its high. All down to volatility decay I guess ?
UPRO iS 3X S&P500. As the S&P500 fell more than 33.4%, the rebalancing actually saved it. Otherwise it would be worth 0.
RocketShipTech
Posts: 679
Joined: Sat Jun 13, 2020 10:08 pm

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

Post by RocketShipTech »

Zb3 wrote: Fri Jun 19, 2020 5:03 pm Anyone notice that the s&p 500 is 10 percent off it’s all time high but UPRO is 80 percent off its high. All down to volatility decay I guess ?
Huh? UPRO is only 45% down from its high...
Rad Dad
Posts: 30
Joined: Fri Aug 09, 2019 9:48 pm

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

Post by Rad Dad »

langlands wrote: Sat Jun 20, 2020 3:19 pm
In terms of the literal mechanics of implementing the trade, adding VIX calls into your portfolio probably isn't much more complicated than adding SPX puts into your portfolio considering both securities are readily available at most brokerages. I'm more concerned about the fact that VIX calls are a "2nd order" instrument whereas SPX puts are only a "1st order" instrument (and the SPX index would be a "0th order" instrument). What I mean by that is that to use SPX puts, you need to understand the Black-Scholes formula (which is only itself an approximation) and how volatility is priced, and what a volatility surface is etc. The theory is quite complex mathematically. To understand VIX calls, you need to understand how vol of vol is priced. This brings in even more complexity, and as someone who did an intern project once on stochastic volatility models, this stuff isn't easy and a lot of it is still highly academic.

When I'm constructing my portfolio, thinking about hedging with volatility is complex enough. I'm just skeptical about adding in volatility of volatility itself as an additional complexity to consider.
Thanks for your reply. I understand what you’re saying about VIX being a second order instrument. I guess what appeals to me about using the vix calls is that CBOE (whom I consider a pretty reliable source) has an index that seems relatively simple to reconstruct myself. I probably don’t even need a calculator to figure out how many calls to buy each month.

The SPY puts on the other hand seem to require more calculating. Or at least I’ve yet to see someone suggest a relatively simple buy x amount of puts at y strike and z interval.

Part of the appeal of the Adventure (at least to me) is that it is very low maintenance. I’m not looking to bust out a calculator to go through Black-Scholes. Maybe I’m making it out to be more complicated than it is though.
Semantics
Posts: 168
Joined: Tue Mar 10, 2020 1:42 pm

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

Post by Semantics »

langlands wrote: Sat Jun 20, 2020 3:19 pm
Rad Dad wrote: Sat Jun 20, 2020 2:56 pm
langlands wrote: Sat Jun 20, 2020 11:59 am For people who are considering adding VIX to your portfolio, is there a reason you aren't considering simply adding out of the money put options on SPX? Several posts are talking about VIX calls, which seems way more complicated than an option directly on the SPX.
I initially assumed that implementing VIX calls was going to be really complicated, but after reading this article it doesn’t sound too complicated at all.

http://www.cboe.com/blogs/options-hub/2 ... vxth-index

I’m interested in finding a way to replace a small portion of my TMF allocation with an alternative hedge so I’m more than willing to hear why one is better than the other.
In terms of the literal mechanics of implementing the trade, adding VIX calls into your portfolio probably isn't much more complicated than adding SPX puts into your portfolio considering both securities are readily available at most brokerages. I'm more concerned about the fact that VIX calls are a "2nd order" instrument whereas SPX puts are only a "1st order" instrument (and the SPX index would be a "0th order" instrument). What I mean by that is that to use SPX puts, you need to understand the Black-Scholes formula (which is only itself an approximation) and how volatility is priced, and what a volatility surface is etc. The theory is quite complex mathematically. To understand VIX calls, you need to understand how vol of vol is priced. This brings in even more complexity, and as someone who did an intern project once on stochastic volatility models, this stuff isn't easy and a lot of it is still highly academic.

When I'm constructing my portfolio, thinking about hedging with volatility is complex enough. I'm just skeptical about adding in volatility of volatility itself as an additional complexity to consider.
To get the same effect with SPX puts as something like ^VTXH you'd have to buy very far OTM puts that are very close to expiry, which makes me wonder about liquidity. I think it may require using a larger fraction of the portfolio for the hedge, especially when most of the equity part is leveraged. My reasoning for looking at VIX was just that if I assume 0.5% of the portfolio is going to be burned on an hedge every month, I want that hedge to have high volatility and high convexity, and it seems like you can get more of that with VIX calls than SPX puts.

This article goes into this in more details: https://www.soa.org/globalassets/assets ... -metli.pdf The takeaway is that VIX calls are significantly cheaper for hedging against tail risk, and SPX puts are better for hedging against moderate drops. "If the future repeats the historical average, VIX calls would be more efficient if the market drops by more than 6 percent over the next month."
am
Posts: 3561
Joined: Sun Sep 30, 2007 9:55 am

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

Post by am »

I’m 75% upro and 25% tmf with my quarterly rebalance coming up. Seems like I’m going to sell a ton of upro to buy tmf which doesn’t have much room to gain. Are we still sticking with 55% upro/45% tmf?
langlands
Posts: 612
Joined: Wed Apr 03, 2019 10:05 pm

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

Post by langlands »

Semantics wrote: Sat Jun 20, 2020 5:26 pm
langlands wrote: Sat Jun 20, 2020 3:19 pm
Rad Dad wrote: Sat Jun 20, 2020 2:56 pm
langlands wrote: Sat Jun 20, 2020 11:59 am For people who are considering adding VIX to your portfolio, is there a reason you aren't considering simply adding out of the money put options on SPX? Several posts are talking about VIX calls, which seems way more complicated than an option directly on the SPX.
I initially assumed that implementing VIX calls was going to be really complicated, but after reading this article it doesn’t sound too complicated at all.

http://www.cboe.com/blogs/options-hub/2 ... vxth-index

I’m interested in finding a way to replace a small portion of my TMF allocation with an alternative hedge so I’m more than willing to hear why one is better than the other.
In terms of the literal mechanics of implementing the trade, adding VIX calls into your portfolio probably isn't much more complicated than adding SPX puts into your portfolio considering both securities are readily available at most brokerages. I'm more concerned about the fact that VIX calls are a "2nd order" instrument whereas SPX puts are only a "1st order" instrument (and the SPX index would be a "0th order" instrument). What I mean by that is that to use SPX puts, you need to understand the Black-Scholes formula (which is only itself an approximation) and how volatility is priced, and what a volatility surface is etc. The theory is quite complex mathematically. To understand VIX calls, you need to understand how vol of vol is priced. This brings in even more complexity, and as someone who did an intern project once on stochastic volatility models, this stuff isn't easy and a lot of it is still highly academic.

When I'm constructing my portfolio, thinking about hedging with volatility is complex enough. I'm just skeptical about adding in volatility of volatility itself as an additional complexity to consider.
To get the same effect with SPX puts as something like ^VTXH you'd have to buy very far OTM puts that are very close to expiry, which makes me wonder about liquidity. I think it may require using a larger fraction of the portfolio for the hedge, especially when most of the equity part is leveraged. My reasoning for looking at VIX was just that if I assume 0.5% of the portfolio is going to be burned on an hedge every month, I want that hedge to have high volatility and high convexity, and it seems like you can get more of that with VIX calls than SPX puts.

This article goes into this in more details: https://www.soa.org/globalassets/assets ... -metli.pdf The takeaway is that VIX calls are significantly cheaper for hedging against tail risk, and SPX puts are better for hedging against moderate drops. "If the future repeats the historical average, VIX calls would be more efficient if the market drops by more than 6 percent over the next month."
Thanks for the article Semantics, very interesting read. It seems that they only compare VIX and SPX options at 30 delta though, which isn't very systematic. For the same delta, VIX options are more "convex" than SPX options and so pay off more in tail scenarios. As you say, you could get that more convex payoff as well by buying far OTM SPX puts with delta below 30. I don't know what "close to expiry" means for you. To make things more concrete, I'll just say that I currently own SPY 250 puts expiring October 16, and I'm satisfied with the hedge it's producing.
Ignorant
Posts: 4
Joined: Thu Nov 16, 2017 12:56 am

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

Post by Ignorant »

rascott wrote: Wed Jun 10, 2020 7:59 am
GiveTendies wrote: Tue Jun 09, 2020 10:50 pm
rascott wrote: Tue Jun 09, 2020 8:16 pm Not really the way I look at it. The total equity in the account is roughly $140k.

$70k equity....3x leveraged. So $210k equity exposure. Mainly holding VOO + micro eminis

$70k long duration (roughly 18.5 duration ) 3x leveraged. That's a little tricky to get right. I want something that's similar to $70k in TMF. You can either buy a lot of short term (2 yr).. maybe one Ultra 30 year..... or some combo of contracts throughout the yield curve. For this quarter that came to be one 2yr, one 10 yr and one 30 yr.

Yes that's actually roughly $400k in notional bond exposure (the 2 year is $200k)..... but the duration of the combo is similar to $70k in TMF.

Just calculated this through because I think it is a very smart form of implementation!

Using 1 ZT + 1 ZN + 1 ZB you are sitting at ~530k notional bonds amount with a weighted, adj duration of 6.45 years

This equates to 70k with a duration of 49 years. Unfortunately, Direxion has removed their adj duration on the TMF website (i swear i saw it a few weeks ago) but using 3x TLT from iShares website, we can approximate 3*18.8 = 56.4y duration for TMF.

So you are a little under TMF's exposure but not by much. Using 2 ZT instead would move your duration to 55y and 3 ZT to 61.3y.


I'm curious how you manage your margin/cash in your account? Do you just sit on the remaining cash or do you hold some money-fund like ETF?
I keep some cash just sitting there.... other was in a money like fund such as BIL. With rates at zero might as well move it out of BIL.....I believe that's actually a negative yielding product right now. In an era where cash paid something might matter more.

Have roughly $33k in cash (23% of account). Now that I think about it, as the account value has increased I could probably drop one micro SP500 contract and move $15k or so into VOO.

Somewhere there is a post on here from Sept (in another thread) where another member helped me use the DV01 values of the Treasury futures contracts to find the appropriate contracts for duration I was targeting. Search for DV01 and you'll find it.

I have only a rudimentary knowledge on the subject, but enough to know I'm close enough and it's worked as I expected so far.
Hi Rascott,

As my username suggests, I'm very ignorant but I've read your posts and the one regarding DV01.

I feel like I am almost there but I'm still not quite.

Would you be able to help me an example of how you work out which contracts are required to hit a target leverage. For example, how did you work out that one 2yr, one 10yr, and one 30yr is equivalent to $70K in TMF.
User avatar
physixfan
Posts: 210
Joined: Wed Apr 03, 2019 4:09 pm

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

Post by physixfan »

For all those people who are interested in VIX derivatives... Have you ever seen the performance of VIIX over the last decade? Do you expect to outperform it significantly by rolling call options by yourself, or you think it is acceptable to hold something that’s destined to lose 99% of value over 10 years?
Semantics
Posts: 168
Joined: Tue Mar 10, 2020 1:42 pm

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

Post by Semantics »

physixfan wrote: Sat Jun 20, 2020 10:15 pm For all those people who are interested in VIX derivatives... Have you ever seen the performance of VIIX over the last decade? Do you expect to outperform it significantly by rolling call options by yourself, or you think it is acceptable to hold something that’s destined to lose 99% of value over 10 years?
Yes. Using call options allows for leverage, so that the VIX portion of the portfolio can be confined to a few percent of the portfolio. That makes a big difference to the tax implications of rebalancing, plus you get 60/40 treatment via section 1256.

As to the second question, yes, I believe that holding something that's most likely going to be worthless is acceptable. I do it for my car and my house, so why not my portfolio too?

I'm not sure whether this will be better than holding TMF, but that's why I'm trying both strategies in separate subportfolios.
langlands
Posts: 612
Joined: Wed Apr 03, 2019 10:05 pm

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

Post by langlands »

Semantics wrote: Sun Jun 21, 2020 12:32 am
physixfan wrote: Sat Jun 20, 2020 10:15 pm For all those people who are interested in VIX derivatives... Have you ever seen the performance of VIIX over the last decade? Do you expect to outperform it significantly by rolling call options by yourself, or you think it is acceptable to hold something that’s destined to lose 99% of value over 10 years?
Yes. Using call options allows for leverage, so that the VIX portion of the portfolio can be confined to a few percent of the portfolio. That makes a big difference to the tax implications of rebalancing, plus you get 60/40 treatment via section 1256.

As to the second question, yes, I believe that holding something that's most likely going to be worthless is acceptable. I do it for my car and my house, so why not my portfolio too?

I'm not sure whether this will be better than holding TMF, but that's why I'm trying both strategies in separate subportfolios.
Why not hold both TMF and VIX? It's not like they're highly correlated. Most likely they're both beneficial.
User avatar
physixfan
Posts: 210
Joined: Wed Apr 03, 2019 4:09 pm

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

Post by physixfan »

Semantics wrote: Sun Jun 21, 2020 12:32 am
physixfan wrote: Sat Jun 20, 2020 10:15 pm For all those people who are interested in VIX derivatives... Have you ever seen the performance of VIIX over the last decade? Do you expect to outperform it significantly by rolling call options by yourself, or you think it is acceptable to hold something that’s destined to lose 99% of value over 10 years?
Yes. Using call options allows for leverage, so that the VIX portion of the portfolio can be confined to a few percent of the portfolio. That makes a big difference to the tax implications of rebalancing, plus you get 60/40 treatment via section 1256.

As to the second question, yes, I believe that holding something that's most likely going to be worthless is acceptable. I do it for my car and my house, so why not my portfolio too?

I'm not sure whether this will be better than holding TMF, but that's why I'm trying both strategies in separate subportfolios.
Although VIX derivatives can be only a small portion of your portfolio, most of the time you need to move money from equity to VIX derivatives every time you rebalance. So the bleeding is unlimited.

We buy cars because we need to use them not because they will increase value, but VIX derivatives are not useful in real life, so they are different...
langlands
Posts: 612
Joined: Wed Apr 03, 2019 10:05 pm

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

Post by langlands »

physixfan wrote: Sun Jun 21, 2020 12:37 pm
Semantics wrote: Sun Jun 21, 2020 12:32 am
physixfan wrote: Sat Jun 20, 2020 10:15 pm For all those people who are interested in VIX derivatives... Have you ever seen the performance of VIIX over the last decade? Do you expect to outperform it significantly by rolling call options by yourself, or you think it is acceptable to hold something that’s destined to lose 99% of value over 10 years?
Yes. Using call options allows for leverage, so that the VIX portion of the portfolio can be confined to a few percent of the portfolio. That makes a big difference to the tax implications of rebalancing, plus you get 60/40 treatment via section 1256.

As to the second question, yes, I believe that holding something that's most likely going to be worthless is acceptable. I do it for my car and my house, so why not my portfolio too?

I'm not sure whether this will be better than holding TMF, but that's why I'm trying both strategies in separate subportfolios.
Although VIX derivatives can be only a small portion of your portfolio, most of the time you need to move money from equity to VIX derivatives every time you rebalance. So the bleeding is unlimited.

We buy cars because we need to use them not because they will increase value, but VIX derivatives are not useful in real life, so they are different...
Perhaps insurance is a better example. And that's exactly what the VIX is- a form of insurance. The basic idea is simple: VIX is a negative returning asset that is very negatively correlated to the stock market. From a portfolio optimization perspective, this negative correlation can be valuable enough to overcome its negative return.

Similarly, why do we hold bonds? Because even though it returns less than stocks, it has zero (or even slightly negative) correlation to the stock market and therefore makes the portfolio more efficient (from portfolio optimization perspective). In fact, notice that a 60/40 portfolio of stocks and bonds can be recast as a 100% allocation to stocks plus a -40/40 blend of stocks and bonds. Since stocks return more than bonds on average, this -40/40 portfolio has negative return. And yet we're all ok with holding this negative returning asset.
User avatar
physixfan
Posts: 210
Joined: Wed Apr 03, 2019 4:09 pm

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

Post by physixfan »

langlands wrote: Sun Jun 21, 2020 1:31 pm
physixfan wrote: Sun Jun 21, 2020 12:37 pm
Semantics wrote: Sun Jun 21, 2020 12:32 am
physixfan wrote: Sat Jun 20, 2020 10:15 pm For all those people who are interested in VIX derivatives... Have you ever seen the performance of VIIX over the last decade? Do you expect to outperform it significantly by rolling call options by yourself, or you think it is acceptable to hold something that’s destined to lose 99% of value over 10 years?
Yes. Using call options allows for leverage, so that the VIX portion of the portfolio can be confined to a few percent of the portfolio. That makes a big difference to the tax implications of rebalancing, plus you get 60/40 treatment via section 1256.

As to the second question, yes, I believe that holding something that's most likely going to be worthless is acceptable. I do it for my car and my house, so why not my portfolio too?

I'm not sure whether this will be better than holding TMF, but that's why I'm trying both strategies in separate subportfolios.
Although VIX derivatives can be only a small portion of your portfolio, most of the time you need to move money from equity to VIX derivatives every time you rebalance. So the bleeding is unlimited.

We buy cars because we need to use them not because they will increase value, but VIX derivatives are not useful in real life, so they are different...
Perhaps insurance is a better example. And that's exactly what the VIX is- a form of insurance. The basic idea is simple: VIX is a negative returning asset that is very negatively correlated to the stock market. From a portfolio optimization perspective, this negative correlation can be valuable enough to overcome its negative return.

Similarly, why do we hold bonds? Because even though it returns less than stocks, it has zero (or even slightly negative) correlation to the stock market and therefore makes the portfolio more efficient (from portfolio optimization perspective). In fact, notice that a 60/40 portfolio of stocks and bonds can be recast as a 100% allocation to stocks plus a -40/40 blend of stocks and bonds. Since stocks return more than bonds on average, this -40/40 portfolio has negative return. And yet we're all ok with holding this negative returning asset.
Yes, insurance is a good analogy. Yet what's important is the premium of the insurance. Bond is positive returning asset, it just has less return than equity; however, VIX derivatives have negative return, and it loses 99% of value in a decade...
User avatar
BB76
Posts: 42
Joined: Mon Dec 17, 2018 3:12 pm

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

Post by BB76 »

physixfan wrote: Sun Jun 21, 2020 2:43 pm
langlands wrote: Sun Jun 21, 2020 1:31 pm
physixfan wrote: Sun Jun 21, 2020 12:37 pm
Semantics wrote: Sun Jun 21, 2020 12:32 am
physixfan wrote: Sat Jun 20, 2020 10:15 pm For all those people who are interested in VIX derivatives... Have you ever seen the performance of VIIX over the last decade? Do you expect to outperform it significantly by rolling call options by yourself, or you think it is acceptable to hold something that’s destined to lose 99% of value over 10 years?
Yes. Using call options allows for leverage, so that the VIX portion of the portfolio can be confined to a few percent of the portfolio. That makes a big difference to the tax implications of rebalancing, plus you get 60/40 treatment via section 1256.

As to the second question, yes, I believe that holding something that's most likely going to be worthless is acceptable. I do it for my car and my house, so why not my portfolio too?

I'm not sure whether this will be better than holding TMF, but that's why I'm trying both strategies in separate subportfolios.
Although VIX derivatives can be only a small portion of your portfolio, most of the time you need to move money from equity to VIX derivatives every time you rebalance. So the bleeding is unlimited.

We buy cars because we need to use them not because they will increase value, but VIX derivatives are not useful in real life, so they are different...
Perhaps insurance is a better example. And that's exactly what the VIX is- a form of insurance. The basic idea is simple: VIX is a negative returning asset that is very negatively correlated to the stock market. From a portfolio optimization perspective, this negative correlation can be valuable enough to overcome its negative return.

Similarly, why do we hold bonds? Because even though it returns less than stocks, it has zero (or even slightly negative) correlation to the stock market and therefore makes the portfolio more efficient (from portfolio optimization perspective). In fact, notice that a 60/40 portfolio of stocks and bonds can be recast as a 100% allocation to stocks plus a -40/40 blend of stocks and bonds. Since stocks return more than bonds on average, this -40/40 portfolio has negative return. And yet we're all ok with holding this negative returning asset.
Yes, insurance is a good analogy. Yet what's important is the premium of the insurance. Bond is positive returning asset, it just has less return than equity; however, VIX derivatives have negative return, and it loses 99% of value in a decade...
I don't mean to step in over my head as I'm very grateful to be able to read such thorough and thoughtful discussions. Perhaps an irrelevant comment / question, but to continue the insurance analogy ... paying premiums can seem like an unnecessary expense (negative return) for years, if not the entire life of a policy. However, you're certainly glad you have the coverage if needed in an emergency. My understanding of the discussion is that a small inclusion of VIX could provide such a "policy." Am I understanding the basis of the argument for including VIX exposure correctly?
“Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius — and a lot of courage to move in the opposite direction.” ― E.F. Schumacher
Rad Dad
Posts: 30
Joined: Fri Aug 09, 2019 9:48 pm

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

Post by Rad Dad »

physixfan wrote: Sun Jun 21, 2020 2:43 pm
Yes, insurance is a good analogy. Yet what's important is the premium of the insurance. Bond is positive returning asset, it just has less return than equity; however, VIX derivatives have negative return, and it loses 99% of value in a decade...
I think part of the interest in VIX derivatives is that there is some doubt that bonds are going to be a positive returning asset in the near future. I’m not totally convinced that bonds are trash going forward like some are saying, but I think it’s enough of a possibility to at least consider a second hedge. Especially one that only accounts for a very small amount of my portfolio.
ChrisBenn
Posts: 453
Joined: Mon Aug 05, 2019 7:56 pm

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

Post by ChrisBenn »

Rad Dad wrote: Sun Jun 21, 2020 3:57 pm
physixfan wrote: Sun Jun 21, 2020 2:43 pm
Yes, insurance is a good analogy. Yet what's important is the premium of the insurance. Bond is positive returning asset, it just has less return than equity; however, VIX derivatives have negative return, and it loses 99% of value in a decade...
I think part of the interest in VIX derivatives is that there is some doubt that bonds are going to be a positive returning asset in the near future. I’m not totally convinced that bonds are trash going forward like some are saying, but I think it’s enough of a possibility to at least consider a second hedge. Especially one that only accounts for a very small amount of my portfolio.
But it's 100% certain that VIX won't be a positive returning asset, so seem like me to be a dubious tradeoff if TMF future return is your concern :)
Rad Dad
Posts: 30
Joined: Fri Aug 09, 2019 9:48 pm

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

Post by Rad Dad »

ChrisBenn wrote: Sun Jun 21, 2020 4:56 pm
Rad Dad wrote: Sun Jun 21, 2020 3:57 pm
physixfan wrote: Sun Jun 21, 2020 2:43 pm
Yes, insurance is a good analogy. Yet what's important is the premium of the insurance. Bond is positive returning asset, it just has less return than equity; however, VIX derivatives have negative return, and it loses 99% of value in a decade...
I think part of the interest in VIX derivatives is that there is some doubt that bonds are going to be a positive returning asset in the near future. I’m not totally convinced that bonds are trash going forward like some are saying, but I think it’s enough of a possibility to at least consider a second hedge. Especially one that only accounts for a very small amount of my portfolio.
But it's 100% certain that VIX won't be a positive returning asset, so seem like me to be a dubious tradeoff if TMF future return is your concern :)
I wouldn’t consider replacing all of the TMF allocation. Just a small percentage.

I’m assuming that the driving factor in future returns will be UPRO. I’m dubious of TMFs ability to duplicate the returns of the past decade. To me both TMF and VIX are essentially crash insurance. I think TMF can still do that, but if rates start increasing over a period of several years my 45% TMF allocation will get hammered. If allocating a small amount of TMF space to VIX allows me a greater increase in UPRO that’s something I’m interested in investigating.

Essentially this comes down to the same concerns that initially caused the UPRO allocation to increase from 40% to 55%.

I’m not really sure if using VIX derivatives would really work out that way so I’m interested in others opinions.
Semantics
Posts: 168
Joined: Tue Mar 10, 2020 1:42 pm

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

Post by Semantics »

physixfan wrote: Sun Jun 21, 2020 12:37 pm
Semantics wrote: Sun Jun 21, 2020 12:32 am
physixfan wrote: Sat Jun 20, 2020 10:15 pm For all those people who are interested in VIX derivatives... Have you ever seen the performance of VIIX over the last decade? Do you expect to outperform it significantly by rolling call options by yourself, or you think it is acceptable to hold something that’s destined to lose 99% of value over 10 years?
Yes. Using call options allows for leverage, so that the VIX portion of the portfolio can be confined to a few percent of the portfolio. That makes a big difference to the tax implications of rebalancing, plus you get 60/40 treatment via section 1256.

As to the second question, yes, I believe that holding something that's most likely going to be worthless is acceptable. I do it for my car and my house, so why not my portfolio too?

I'm not sure whether this will be better than holding TMF, but that's why I'm trying both strategies in separate subportfolios.
Although VIX derivatives can be only a small portion of your portfolio, most of the time you need to move money from equity to VIX derivatives every time you rebalance. So the bleeding is unlimited.

We buy cars because we need to use them not because they will increase value, but VIX derivatives are not useful in real life, so they are different...
I was referring to insurance, which maybe is better suited for a tail risk hedging thread, but it seems relevant here too since 3x leverage means a sharp correction can look more like a black swan.

There's also the low bond yields problem. Since 2010, TLT has been negative in three years, and in all three 55 UPRO/45 TMF underperformed SPX, with similar results on a monthly basis. So going forward the basic UPRO/TMF strategy might not perform as we hope. I have no plans to stop holding treasuries, but don't want to incur much more risk than an unleveraged portfolio in order to barely outperform it. Finding a better reward/risk tradeoff probably requires either leveraging less, or finding something else to mix in.

A bit of ^VIX mixed in seems to help most portfolios in backtesting. The rub is that it's expensive to track VIX, but maybe in the short term when volatility is high the futures being in backwardation can help make this cheap -- VIXY has "outperformed" ^VIX since March.

https://www.portfoliovisualizer.com/bac ... ion2_2=100
jolmscheid
Posts: 78
Joined: Fri Feb 24, 2017 1:22 pm

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

Post by jolmscheid »

Why not look more into a target volatility model for protection instead of VIX? PV shows that it could be beneficial: https://www.portfoliovisualizer.com/tes ... ation3_1=0
jolmscheid
Posts: 78
Joined: Fri Feb 24, 2017 1:22 pm

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

Post by jolmscheid »

Or, a moving average model in attempt to lower drawdowns/risk: https://www.portfoliovisualizer.com/tes ... odWeight=0
Semantics
Posts: 168
Joined: Tue Mar 10, 2020 1:42 pm

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

Post by Semantics »

The timing models are an interesting twist, I'm using inverse volatility because it's simple, seems to drive down volatility more than target volatility in practice, and is somewhat more friendly to a taxable account (no months where it goes 100% UPRO or 100% TMF like the others). It works best if the assets all have similar volatilities, otherwise it'll just put a heavy weight on the lowest one, even if the returns are also low.

One other thing worth mentioning: TMF's annual return was only 1.53% from 2012-2018 as the 10 year ended up about +0.5 during that interval. So that may be a good period to backtest on for those worried about poor bond returns going forward. The good news is 55/45 still does well. The Sharpe ratio is in line with that of SPX, which makes sense since there are no free returns from rate cuts. And the adaptive allocation models do very well in that environment too. I'm less interested in using VIX after seeing these data.
jolmscheid
Posts: 78
Joined: Fri Feb 24, 2017 1:22 pm

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

Post by jolmscheid »

Semantics wrote: Sun Jun 21, 2020 8:45 pm The timing models are an interesting twist, I'm using inverse volatility because it's simple, seems to drive down volatility more than target volatility in practice, and is somewhat more friendly to a taxable account (no months where it goes 100% UPRO or 100% TMF like the others). It works best if the assets all have similar volatilities, otherwise it'll just put a heavy weight on the lowest one, even if the returns are also low.

One other thing worth mentioning: TMF's annual return was only 1.53% from 2012-2018 as the 10 year ended up about +0.5 during that interval. So that may be a good period to backtest on for those worried about poor bond returns going forward. The good news is 55/45 still does well. The Sharpe ratio is in line with that of SPX, which makes sense since there are no free returns from rate cuts. And the adaptive allocation models do very well in that environment too. I'm less interested in using VIX after seeing these data.
For inverse volatility, is there a volatility period / timing period that seems to bode well with the strategy?
Semantics
Posts: 168
Joined: Tue Mar 10, 2020 1:42 pm

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

Post by Semantics »

jolmscheid wrote: Sun Jun 21, 2020 8:57 pm
Semantics wrote: Sun Jun 21, 2020 8:45 pm The timing models are an interesting twist, I'm using inverse volatility because it's simple, seems to drive down volatility more than target volatility in practice, and is somewhat more friendly to a taxable account (no months where it goes 100% UPRO or 100% TMF like the others). It works best if the assets all have similar volatilities, otherwise it'll just put a heavy weight on the lowest one, even if the returns are also low.

One other thing worth mentioning: TMF's annual return was only 1.53% from 2012-2018 as the 10 year ended up about +0.5 during that interval. So that may be a good period to backtest on for those worried about poor bond returns going forward. The good news is 55/45 still does well. The Sharpe ratio is in line with that of SPX, which makes sense since there are no free returns from rate cuts. And the adaptive allocation models do very well in that environment too. I'm less interested in using VIX after seeing these data.
For inverse volatility, is there a volatility period / timing period that seems to bode well with the strategy?
I've been backtesting with 1 month (20 trading days) and it seems to work well enough, but I'm not sure whether it's optimal. In a volatile market like this I figure it's better to be rebalancing relatively often.
RayKeynes
Posts: 259
Joined: Mon Nov 11, 2019 2:14 am

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

Post by RayKeynes »

@ HedgeFundie or @ all: What is the historically best split-up between TMF and UPRO? I think its around 40/60 (TMF/UPRO), or am I wrong?

I was thinking of going a little bit more "agressive" with a UPRO/TMF approach of 80/20 above SMA235 abd with 45 / 55 below SMA235. What do you guys think?
occambogle
Posts: 592
Joined: Thu Dec 12, 2019 4:58 am

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

Post by occambogle »

FYI for people into VIX.... Credit Suisse are saying TVIX will be delisted:

https://www.reuters.com/article/us-cred ... SKBN23T2BI
User avatar
Meaty
Posts: 796
Joined: Mon Jul 22, 2013 7:35 pm
Location: Texas

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

Post by Meaty »

RayKeynes wrote: Mon Jun 22, 2020 9:47 am @ HedgeFundie or @ all: What is the historically best split-up between TMF and UPRO? I think its around 40/60 (TMF/UPRO), or am I wrong?

I was thinking of going a little bit more "agressive" with a UPRO/TMF approach of 80/20 above SMA235 abd with 45 / 55 below SMA235. What do you guys think?
Recommend you look at PV. 55/45 UPRO/TMF gives the highest sharpe ratio
"Discipline equals Freedom" - Jocko Willink
jolmscheid
Posts: 78
Joined: Fri Feb 24, 2017 1:22 pm

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

Post by jolmscheid »

RayKeynes wrote: Mon Jun 22, 2020 9:47 am @ HedgeFundie or @ all: What is the historically best split-up between TMF and UPRO? I think its around 40/60 (TMF/UPRO), or am I wrong?

I was thinking of going a little bit more "agressive" with a UPRO/TMF approach of 80/20 above SMA235 abd with 45 / 55 below SMA235. What do you guys think?
I like the idea of utilizing a moving average of some sort in an attempt to cut down on the more drastic drawdowns.
jolmscheid
Posts: 78
Joined: Fri Feb 24, 2017 1:22 pm

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

Post by jolmscheid »

So, UPRO/TMF re-leverage daily. With the circuit breaker rules in place, the market can drop by a Max of 20% in a trading day. This would equate to a 60% drop for the 3x leveraged fund(s), if this were to happen...so, with re-leveraging daily, there is virtually no way for the 3x leveraged funds to go to zero.

Not saying that we want this to happen, but for this to go "bust" the market and long term Treasuries would both have to drop over 15% in one day. Is this correct? So, when looking at PV, and the S&P shows a Max drawdown for example of -50% let's say, this strategy would still be "alive" due to the daily re-leveraging?
Lee_WSP
Posts: 3465
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

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

Post by Lee_WSP »

jolmscheid wrote: Mon Jun 22, 2020 6:24 pm
RayKeynes wrote: Mon Jun 22, 2020 9:47 am @ HedgeFundie or @ all: What is the historically best split-up between TMF and UPRO? I think its around 40/60 (TMF/UPRO), or am I wrong?

I was thinking of going a little bit more "agressive" with a UPRO/TMF approach of 80/20 above SMA235 abd with 45 / 55 below SMA235. What do you guys think?
I like the idea of utilizing a moving average of some sort in an attempt to cut down on the more drastic drawdowns.
You would've been whipsawed in '18 & '20.
User avatar
cos
Posts: 235
Joined: Fri Aug 23, 2019 7:34 pm
Location: Boston
Contact:

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

Post by cos »

jolmscheid wrote: Tue Jun 23, 2020 9:34 pm So, UPRO/TMF re-leverage daily. With the circuit breaker rules in place, the market can drop by a Max of 20% in a trading day. This would equate to a 60% drop for the 3x leveraged fund(s), if this were to happen...so, with re-leveraging daily, there is virtually no way for the 3x leveraged funds to go to zero.

Not saying that we want this to happen, but for this to go "bust" the market and long term Treasuries would both have to drop over 15% in one day. Is this correct? So, when looking at PV, and the S&P shows a Max drawdown for example of -50% let's say, this strategy would still be "alive" due to the daily re-leveraging?
Even if both the S&P 500 and long-term US treasuries drop 15% in one day, you won't necessarily go bust. They'd both have to drop 34%, and as you said, that's impossible for UPRO given the existence of circuit breakers monitoring the S&P 500 and unlikely for TMF given the stability of TLT (its largest recorded 1-day loss in over 18 years was -6.43%). It's also unlikely that they'd both drop so steeply simultaneously given their fairly large degree of anticorrelation.

The drawdowns reported by Portfolio Visualizer happen over the course of multiple months or even years, not single days, so yes, this strategy would live thanks to both its daily releveraging and its quarterly rebalancing. For example, during the subprime mortgage crisis, the S&P 500 suffered a -50.97% drawdown while this strategy suffered somewhere in the ballpark of a -67% drawdown.

You should take another look at the data in the first post of the original thread. A lot of your questions can be answered by exploring it yourself, and you'll probably learn something interesting in the process! It's really cool stuff.
Post Reply