Most of us live in a linear world.
Markets plunge and we rely on our wits to jump out at the right time. More often than not you fuck it up, because here's the thing, it's impossible to consistently get right.
Sure you can have natural hedges which reduce risk exposure just as markets plunge, but they are only breadcrumbs of consolation as your last supper is snatched away from you.
Sure timing strategies may get lucky - there is little doubt they were 'unlucky' during backtesting, put it that way!
The ultimate backtest is the future and the future's a bitch.
'Tempus fugit' as they say. Time flies during a crisis.
What can you do?
Hedge your tail risks and then stress test (even most professionals do a very poor job at this).
There is a point beyond which every portfolio cannot come back from, be it bankruptcy or getting divorced.
1. Set a maximum allowed loss.
So, say for example you have devoted your life savings to the S&P 500, you could visualise a loss limit of 50%, like thus
How do we ensure that we never cross that 50% line? We can hedge with a 50% out of the money put option.
Now we cannot lose more than 50% and can rest assured that our institution won't go bankrupt.
Let's change our visual symbology a little.
Shaded areas now mean apocalyptic danger and if we hedge away all the danger we see no shaded areas.
Of course even the simplest portfolios have more than one 'risk factor'.
2. Identify pertinent risk factors
Be those stocks by sector, e.g. perhaps the big risk factors in your portfolio are energy stocks and tech? Foreign and domestic stocks. FX or credit spread risk.
Let's look at two dimensions.
This is an option portfolio which contains just a single call option.
Now we have two dimensions. Equity price and volatility.
The downward sloping line is the line which seperates danger from safety. Perhaps this line represents the loss of 40% or so, anything below that may cause margin calls etc.
The shaded area means large equity or volatility price drops will bankrupt us.
How would we go about removing this danger?
Could we remove it?
Or would we even want to remove it completely?
First thing that comes to mind is writing a call (with a strike greater than initial call).
When you write a call you get cash in return for the option.
On the one hand you are more liquid, on the other if equity prices or volatility increases a lot you might get screwed depending on how large a call you wrote.
Now you can be screwed with large movements in either direction, but this might be acceptable if you began by being very close to imminent danger and get more breathing space from the option's premium.
Which leads to the next step
3. Set maximum possible shocks
Start by looking at historical shock magnitudes over your time horizon and add a percentage for safety
In the example above we have equity and vol dropping by 50%. Obviously after writing calls we should also set shock jump limits too!
This implicitly acknowledges that there may be some shocks which we will never be able to survive, but when aliens invade your job as a risk manager is done and you should take out your shotgun.
Nevertheless if you see any shaded areas on your map, you should go about removing them.
You have identified (1) a line in the sand; (2) the important risk factors (dimension reduction always plays a major role in this type of analysis); and (3) maximum possible shocks.
Now you can draw a 'Stress Map' for your portfolio and defend your it like you'd play Kingdom Rush!
Note we haven't touched anything statistical, because statistics won't save you in a crisis.