Simple stock bond strategy hedged using VIX options

Hi Narendra, 

thanks for sharing the backtest and idea. From my perspective, it often makes sense to add a long (fwd) vol component to most traditional asset allocations. You are probably aware of it, but there are quite a few hedge funds and alternative asset managers out there that pitch this underlying idea. For example, one of the most vocal proponents of long vol exposure is Chris Cole from Artemis. You can check out their research on their website. Could be interesting for you if you have not already read it. They have an interesting view when it comes to volatility investing and portfolio allocation.

Best,
Adrian

Hey Adrian,

I completely agree with you. Issue is how to you add a cheap long volatility component? I tried to add a long volatility component by buying cheap OTM VIX options. But its not “active” in the sense we always buy long volatility. If there was a signal then we could perhaps do better. DO you have any suggestions on now to add a long volatility component that does not have a large drag on the returns.

Narendra Kulkarni You are probably aware of In-and-Out concept discussed on QC. I personally like Vladimir  V1.3 version discussed in “Intersection of ROC Comparison using out day approach”. Link here. Algorithm tries to predict whether it is risky to be in market or not. It would be interesting to see if you can modify your VIX options hedging depending on the signal from V1.3.

Narendra Kulkarni I agree that it is difficult to find long volatility components without too much bleeding. This is a topic that I am currently looking into myself. However, I just started using QuantConnect more actively and still have to get familiar with the framework. So far I did not find sth. that I am truly happy with. In my mind, there are two ways this issue can be approached (pls let me know what you think):

• Combination with signals (= dynamic portfolio allocation / timing)
  • Entry signal
    • One could argue that your option weighting is based on a signal (in your case the absolute level of VIX)
    • Any other potentially meaningful signal… (e.g. vol term structure?)
  • Exit signal
    • Trade-off decision between continuous vol smoothing (keeping long vol component until expiry in your case) and taking profit (likely at suboptimal time and, thereby, missing out on some further vol moves)
  • Generally, one has to be extremely careful when it comes to timing. In my opinion, a lot of signals are fake and result in overfitting (as you know, separate research and validation etc. helps to mitigate this to some extend)
• Underlying long vol strategy (just some quick ideas)
  • Custom exposure via vix options & futures (or combination thereof)
  • Replication of OTC derivatives via spx options & futures (or combination thereof)
  • Custom exposure via spx options & futures (or combination thereof)
  • Index vol vs. single stock vol
  • However, active risk management might be quite complex which in turn might increase the probability of overfitting

With regards to timing - the underlying question is whether one believes that these extreme moves or volatility spikes are predictable. If one believes they are not predictable at all, then timing does not add value and one might want to have a long volatility component in the portfolio at all times → focus on finding efficient long vol strategies. If one believes volatility spikes are somewhat predictable (due to e.g. autocorrelation, clustering or whatever) then one really has to think about signals. Of course, a mix of both approaches might yield promising results too. 

Not sure, whether that reply is much help – but happy to discuss and hear your views. 

I agree with all of that. Using spy puts and vx futures is where I would look first.

Years ago I created multiple VIX timing entries and exits for options, but they ended up not actually trading well due to the vix options following the vx futures. OTM SPY puts may be a better product to hedge with for a downturn. VX futures may trade at a premium when VIX is low, and trade at a discount when VIX is high, further eroding your call value and lessening the hedge.

However if direct volatility exposure is what you want then you have vix options, vx futures and vix etfs. But you may need a timer to switch between long and short volatility, as always long vx futures will erode to much capital. Have you tried vix straddles?

For the futures I'm currently looking at shorting vx and es futures together when vx futures peak, the vx downturn will come and assumption is that vx futures will drop more than es rises thereby showing a profit.

One way to make any long option cheaper is to sell a spread closer to the money. Yes you lose the requirement if it closes past your spread, but it makes the further out of the money option cheaper to buy outright. that may be something to try if you want to continue vix options, or maybe even too with spy options. For the spy options that would decrease the cost of the OTM puts.

Just some thoughts as I'm currently looking into different ways of hedging my own long positions more effectively.