hedging a portfolio against a black swan

OptionGuru said:
US DOLLARS



Is the best way to prepare for a "black swan".


:)
More...
bjw said:
Hi,

Quick question: imagine you would want to hedge a portfolio, with lots of well-diversified positions consisting of stocks and derivatives, against a "black swan" type of event. The portfolio currently generates some positive returns if the underlying stocks (market in general) goes down 10-20%, but from 25-30% downwards it pretty much turns ugly quite fast.

How do you protect yourself against that? Obviously, by just buying puts on, say, SPY, you'd likely over-pay for insurance as you don't need insurance (on the contrary) for the first 20% down, which is what you're in fact getting. Are there any other option strategies or financial instruments that would do a better job?

Thanks for reading.
More...
BJW

We have developed and back tested a strategy with key2options using low Delta VIX Long Calls specifically to protect against Black Swan events. Here are the rules and parameters and results. This, of course, is not investment advice.

Our back testing model entered a trade on the first day of a quarter and took it off on the last day of the same quarter. It repeated each quarter. It bought calls with 90 DTE and 1-10 Delta (way OTM).

Back testing back to 2007 yields a very few large returns and some smaller ones. Using higher Delta calls did not work well.

It is important to accept that this is not an investment, merely catastrophe insurance, somewhat akin to homeowner's insurance, i.e., asset protection.

We will use other forms of market timing to move to cash when needed to reduce draw downs but most of those methods take time to trigger. Because this insurance is in play 100% of the time, it does not take time to trigger.

While our back testing indicates that this plan does make money, that's not its real purpose which is to protect our primary market assets from Black Swan events.

Before beginning our back testing, we were willing to spend 2% of assets annually (or 1/2% quarterly) for this protection, but the back testing demonstrated that the strategy did indeed make money over the 9.5 year testing window and therefore the insurance was free. It would not have been free if there had been no Black Swans such as the Lehman Brothers bankruptcy.

Bob
 
RLT:
Interesting. Can you clarify your statement: "While our back testing indicates that this plan does make money, that's not its real purpose which is to protect our primary market assets from Black Swan events."? I'm guessing, this means it DID more than adequately protect the "set of assets for which it was intended" and the "does make money" reference implied your long call was actually a higher delta than necessary? (I want to insure, the "does make money" statement was NOT referring to this trade in isolation to the portfolio to which it was meant to protect)
 
stepandfetchit said:
RLT:
Interesting. Can you clarify your statement: "While our back testing indicates that this plan does make money, that's not its real purpose which is to protect our primary market assets from Black Swan events."? I'm guessing, this means it DID more than adequately protect the "set of assets for which it was intended" and the "does make money" reference implied your long call was actually a higher delta than necessary? (I want to insure, the "does make money" statement was NOT referring to this trade in isolation to the portfolio to which it was meant to protect)
More...

You guess correctly. It did make money but I have modeled several other option based systems that were much more effective as investments but they were not in trades 100% of the time as insurance must be (see exception below). That dictates that run ups must be left in play until a quarter's end even at the cost of the run up falling or even becoming a draw down.

I could not have done this testing without the use of a modeling tool as there are too many independent variables to attack using manual methods such as spreadsheets. For example, one of the 38 trades returned over 10,000% in 2008. The "Rules" state that that trade would be left in play, but I modeled to take it as a realized gain because I knew that I would not have left $54,000 on the table if and when it really happened. The point is that there are personal preferences that must be modeled in and it takes the right tools to do so.

I have no financial interest in any of the several companies I use to create mechanical trading systems.

It was important for me to accept that this insurance is simply a cost of doing business as a private trader and the gains it made simply achieved the goal of offsetting market draw downs.

I don't view it as a market timing method, such as SectorSurfer's StormGuard Armor, but more as a complement to a primary market timing strategy.

Again, I have no financial interest in any of the several companies I use to create mechanical trading systems. There is a group to which I belong in Dallas devoted to developing such systems for our own private, non-commercial use.

Bob
 
Someone at ET suggested I could hedge my position with a no cost collar to protect against a black swan. It worked when I was already profitable and did not mind being called away. In this scenario, the key is selecting the collar and width.
 
Dolemite said:
Didn't seem like a very big move in USD compared to the move down in the market. How would you play this as a hedge?
More...


The OP wants to hedge/protect his portfolio from "black swans". My advice is to buy US dollars, ideally by selling part (20% perhaps) of his portfolio for USD. Any action has to be taken before the black swan event.




:)
 
OddTrader said:
" Any action has to be taken before the black swan event. " :)
More...


That's correct, the OP can't procrastinate. Considering the timing of his post he obviously is concerned about a big market correction.


  • Was today the signal to take action?
  • If he waits too long he might end up selling at the market bottom.
  • Decisions, decisions, decisions.


IMO ..... Selling 20% of the portfolio for USD would be a good idea.


:)
 
You must log in or register to reply here.
Back
Top