Trading Black Swans in Forex - page 2

 
jjc:

Sticking somewhat to the original topic, Taleb has some interesting criticisms of VaR, which is a component of RAROC. (However, you have to get beyond the fact that he's got a twitchy finger on the rhetorical trigger. "I have always held that VAR is charlatanism" is a sample quote from his website at https://www.mql5.com/go?link=http://www.fooledbyrandomness.com/quant.html)

It certainly doesn't appear to make much sense if you try to do it with spot forex. You want something much more asymmetric - such as the baskets of options which Gladwell describes Taleb as trading.


VaR, and any risk metric that relies on historical data to compute future risk, will always have limitations to its relevance of application. It is an inescapable consequence of the math involved.

At risk of stating the obvious myself, I personally feel that anyone who feels compelled to make a living admonishing risk metrics for doing nothing more than that which they are intrinsically capable is tantamount to simply stating the obvious.

VaR has its share of mathematically assured weaknesses, but the biggest failures of VaR typically come down to a failure to properly implement VaR. You can't claim VaR is flawed when you run your bank into the ground doing things that intentionally flout meaningful risk assessment (I'm looking at you Dick Fuld).


ubzen:

Omg.. from Black Swans to Black Scholes (just when I was thinking about it). You'd think you're training to become a Kong-fu master. I'm betting you advanced guys know the formula off the top of you heads. What's the big deal with this Soros guy anyway? You'd think he invented trading. Break England's bank, big deal. Guy just got lucky he didn't go broke in the process. Yep, I said it. (Oh-boy they gonna kick me off the site for sure now.)

Soros did exactly what any successful trader does so that they can live to tell about it...he knew when to exit the game. And I don't mean just on that one multi-billion dollar trade. The guy knew when his time was up and a reversion to mean was coming. Gamblers, the profitable ones, know when the table hits this point. As does any criminal that never gets caught.

Its the one's that have a fantastic trading year, an awesome run on the table, made that one killer bank heist, but just couldn't quit the game while ahead and instead they kept trading/playing/heisting in an attempt to relive the glory again that eventually lose it all. The swan song :P
 
badi:

What is a Black Swan :

What it does basically is making small losses and small profits until it catches a long trend.

There are two certainties about this system :

- It makes losses
- It catches big trends

This is essentially the same idea that is behind my snowball anti-grid EA. (google: snowball, anti-grid). It does essentially the opposite of what most losing traders do: It will not take small wins and wait for the big loss, instead it will take small losses and wait for the big profit. Snowball works semiautomatic, you have to decide when to take profit, it still needs discretion. Profits grow quadratically with price, losses grow linear with time. I have phenomenal results with this approach, it is almost unreal.
 

7bit: ...it is almost unreal.

Do you have any back-test results to tease me with?

Edit*Nice site 7bit: http://sites.google.com/site/prof7bit/snowball 

Humm... Looks like a Manual EA.. So how about some forward test results?

 
ubzen:

7bit: ...it is almost unreal.

Do you have any back-test results to tease me with?

Edit*Nice site 7bit: http://sites.google.com/site/prof7bit/snowball

Humm... Looks like a Manual EA.. So how about some forward test results?

http://prof7bit.mt4stats.com/analyses Ignore all USDCHF trades, this means subtract roughly 5000$ profit that result from these usdchf trades, the rest is anti-grid. I use the conventional grid to somehow compensate the ongoing losses between the black swan, this grid is also controlled discreationary and has emergency stops, It is a very special grid implementation that can be paused and reversed at any time. There is also a thread on forexfactory about the snowball EA with equity plot screenshots.
 
1005phillip:
At risk of stating the obvious myself, I personally feel that anyone who feels compelled to make a living admonishing risk metrics for doing nothing more than that which they are intrinsically capable is tantamount to simply stating the obvious.

Taleb's argument is a little more elegant than that. In essence, he's saying that there is a dangerous disparity between what VaR is actually "intrinsically capable" of, versus what it is believed to be "intrinsically capable" of and what it's used for.

You seem to be suggesting that risk metrics such as VaR are limited, but fine within those limitations. Taleb may or may not be right, but he's saying something different. He's definitely not "stating the obvious".

 
ubzen:
Thank you for explaining this system. This sure would have made a good Article. Posting here like this it's gonna get allot of negative comments. I'm not sure if that's a good or bad thing; depends on what you're trying to accomplish. Well it's not a revolutionary concept. Sound like good old trend following to me. For those people not familiar with such a system, you did a good job introducing it to them. Cuddles !!

I appreciate all comments anyway, thanks for backup !
 
EuroTrader:

Chinese torturers has a name for it. Death By A Thousand Cuts. You definitely need a very very deep pocket and finally a therapist to get you out of depression.


Risk percentage is very low so it is very unlikely for this system to make you loose all your money all your money.

Remember, there are two ways of loosing money : Bleed or Blowup, may be you prefer to fool yourself with short term results and wait for the blowup that can make you wish you just spent that money instead of investing it.

 
1005phillip:

why would anyone wait out the 5-7 years it takes for a single black swan event to occur just to monopolize on that once in a decade rare opportunity?

You want to build in safeguards (circuit breakers, just as the establishment uses) to minimize the damage black swans can cause, but to create a strategy that attempts to truly wait out the markets in hopes of catching the next one in the coming decade just seems like a fool's errand to me.

And Eurotrader is right, this system appears no different than a trend-trader that tries to catch the big moves while bleeding during the periods of sideways trading. You don't need to call them black swans.


Again, may be you prefer short term results that make you blowup.

 
1005phillip:

Trend following is real and people make lots of money invoking trend following strategies. The reason why people still opt to pursue alternative trade strategies has to do with maximizing their rate of return (ROR) and/or minimizing their risk of loss (ROL).

Consider that if all any of us were interested in was simply making more money from our current equity then we'd just buy a government backed bank CD and sit back to drink margarita's all day.

What compels us to this industry is the desire to make even more money from our current equity.

Ideally all of us are busy at work devising cleverer and cleverer trading schemes to maximize our Risk Adjusted Return on Capital (RAROC).


Believing too much in systems that give astonishing results (short term results) can get you banckrupt. It is more about risk taking and approach to loss.

Trying to catch the falling knife...i.e. constantly shorting an otherwise bullish trending market in the hopes of catching a once-in-a-decade fundamental market reversal (typically progenitored by the black swan event) does not strike me as a superior RAROC approach.

If you look at the statements you will see that trades are almost 50/50 between Bullish and Bearish
 
1005phillip:

Soros did exactly what any successful trader does so that they can live to tell about it...he knew when to exit the game.
There lays the difference between true investing and gambling, thank you for using that particular phrase.
Reason: