Black swan

Discussions about the testing and simulation of mechanical trading systems using historical data and other methods. Trading Blox Customers should post Trading Blox specific questions in the Customer Support forum.
marriot
Roundtable Knight
Roundtable Knight
Posts: 353
Joined: Thu Nov 20, 2008 3:02 am

Black swan

Post by marriot » Tue Apr 27, 2010 1:18 pm

You have tested, stressed every single corner of your suite.
Then here we are: it happen what we will never like to see, something that in our heart we will never understand.
What you do if you are on the wrong side of market?

stamo
Roundtable Knight
Roundtable Knight
Posts: 522
Joined: Tue Mar 13, 2007 5:27 pm

Post by stamo » Tue Apr 27, 2010 4:32 pm

Execute my plan to get smaller.

Perhaps you could do it in thirds.

1st third often hurts because it's right after a big event and big loss.

2nd third feels better, more rational, a continuing move in the right direction.

After the last 3rd I can look back and figure out how to do better.
Last edited by stamo on Thu May 06, 2010 10:38 am, edited 1 time in total.

Moto moto
Roundtable Knight
Roundtable Knight
Posts: 427
Joined: Mon Jun 01, 2009 4:12 am
Location: once again in the UK

Post by Moto moto » Wed Apr 28, 2010 5:03 am

I have always had a rule for what to do when there is an extreme event - eg; a plane into a building.
That rule is halve all existing trades - even if they are profitable. Assess and then probably exit most positions.
This is only on a true black swan event.
In all my trading years, this has only happened a few times while I have been a discretionary trader. Sept 11 2001, End of Oct 1997 Asian crisis for equities, A few specific stock examples of massive unexpected profit warnings.

By having a simple rule like this it means the result is automatic, and I survive another day. Plus while this is hard to test without hindsight, I wish to have this as a rule as part of my trading plan.
(I apply the same rule to any systematic model just for Sept 11 only.)

things like Greece and the the recent downgrading over the last few days.... is not a black swan event, unless you happen to be a financial reporter or broker who is acting like this came out of the blue and was not expected :)

marriot
Roundtable Knight
Roundtable Knight
Posts: 353
Joined: Thu Nov 20, 2008 3:02 am

Post by marriot » Wed Apr 28, 2010 5:39 am

On 11 sept i was not trading.
I coud not get in touch with uk global from italy for 24 hours or more.
This my tread is a follow up thinking about goodness function that average max DD.

kianti
Roundtable Knight
Roundtable Knight
Posts: 335
Joined: Fri May 02, 2003 6:10 am
Location: Florence - Italy

Post by kianti » Thu May 06, 2010 10:35 am

I was trading on 9/11 (in Italy) and I was long DAX futures before the event!
I will never forget the event.
I managed to get out without losses.
At the time, I was quick and didn't waste time thinking in terms of undervalued and overvalued.
At the time I didn't trade in terms of drawdown. When daily maximum loss was hit, that was the time to turn of PC and go and buy me an ice cream (or glass of Chianti):

best regards, as ever

bevok
Full Member
Full Member
Posts: 20
Joined: Tue Oct 20, 2009 5:38 pm
Location: Northland, New Zealand
Contact:

Post by bevok » Fri May 07, 2010 9:26 pm

Does yesterday count as a Black Swan? :-)

stamo
Roundtable Knight
Roundtable Knight
Posts: 522
Joined: Tue Mar 13, 2007 5:27 pm

Post by stamo » Fri May 07, 2010 9:30 pm

Definitely.

Accenture opening the day in $40 range, trading down $5 & then a penny and then back is the epitome of an extreme event.
Last edited by stamo on Sun May 09, 2010 6:54 pm, edited 1 time in total.

Kiwi
Roundtable Knight
Roundtable Knight
Posts: 513
Joined: Wed Apr 16, 2003 1:18 am
Location: Nowhere near

Post by Kiwi » Sun May 09, 2010 12:51 am

bevok wrote:Does yesterday count as a Black Swan? :-)
I don't really thing it did on any timescale. It was a big move but it was signaled at multiple levels.

On daily we'd been crawling up to dow 11200 and a (bless em) 61.8% retracement of the big move down. Then it did a lower high and collapsed on a traditional head and shoulders pattern.

On the 5m it tried a couple of times at new highs from the overnight session (looking at ES) then it made a lower low and high; and another lower low and high. Then it glissaded down through stop after stop.

That it went so far in one day was a little bit of a surprise but it was definitely on the cards. Currencies have suffered drops related either to underlying issues (and resulting fear) or a retreat from risk (CAD, AUD, NZD) that comes with said fear, uncertainty and doubt.

The really interesting bit becomes: does this presage the big move down that would be a second leg of a truly big bear? Or is it just the retracement that had to happen after a year of running out of the GFC on the back of government handouts around the world. Hmmm ... a bull market based on government handouts?

Soph
Roundtable Fellow
Roundtable Fellow
Posts: 64
Joined: Tue Dec 02, 2008 12:22 pm

Post by Soph » Sun May 09, 2010 10:02 am

bevok wrote:Does yesterday count as a Black Swan? :-)
I guess it depends how you look at it. If you simply look at how the markets closed, I wouldn't think so. Markets have closed on a few significant down days but these certainly weren't unexpected or extreme.

If you include the 1000 point snafu on the Dow and the fact that there are fundamental flaws in the current system to allow such an event to occur, then it likely should be seen as a Black Swan event.

Moto moto
Roundtable Knight
Roundtable Knight
Posts: 427
Joined: Mon Jun 01, 2009 4:12 am
Location: once again in the UK

Post by Moto moto » Sun May 09, 2010 6:12 pm

I would not count it as a black swan event.....
fat fingers, algo trading issues will always occur..... They are not what I would call a life changing/policy changing event, that will cause both immediate and long term severe changes in ideas and perceptions. True black swans are rare. Otherwise it defeats the purpose of the idea?
It does however raise issues of automatic stop losses in a plan.

marriot
Roundtable Knight
Roundtable Knight
Posts: 353
Joined: Thu Nov 20, 2008 3:02 am

Post by marriot » Mon May 10, 2010 1:55 am

When i start to use TB, i realized that i was ovetrading.
I spent almost two years cutting down overal risk.
Last days move kindly suggest me that i have to cut something more.

bevok
Full Member
Full Member
Posts: 20
Joined: Tue Oct 20, 2009 5:38 pm
Location: Northland, New Zealand
Contact:

Post by bevok » Mon May 10, 2010 3:09 am

Although my comment was a bit wry, my take on the Black Swan is that it is an extraordinary event which is thought to be an extreme outlier (typically by people like option writers in the world of the markets) that actually happens more frequently than expected. I don't have the historical data to tell how frequently we should expect a move like Thursday's, however Nassim might argue that those moves are more common than they "should be". Our risk management certainly needs to take them into account. I think in CF's first book he says that risking over 3% would have resulted in wipeout if you were trading Eurodollars during the 1987 crash. We always have to beware of these extreme moves and position size accordingly.

Signature with hyperlink removed -- Moderator

Asamat
Roundtable Knight
Roundtable Knight
Posts: 175
Joined: Fri Jun 03, 2005 7:50 am
Location: Walldorf, Germany

Post by Asamat » Mon May 10, 2010 5:52 pm

However - coming back to the title of the thread - a black swan is not simply a big move or an unexpected move. It is something that before it happened was nearly or totally unthinkable, but which after the fact is so obvious that everybody fools himself into thinking it was to be expected before it happened. Like the real black swans in Australia.

9/11 counts, but a second flight of terrorist planes wouldn't again. 1929 and 1987 count, but similar happenings duplicating those wouldn't. The point Taleb makes with the black swans is that the futures is much more unpredictable and unstructured than we think. Human minds are made to make sense of everything, and so we spin stories explaining and rationalizing things. We "explain" a lot of thing which are just random happenings, and we "explain" black swans, which in reality are unexplainable in the sense that knowledge of one or of several past black swans does not prepare us for the next one.

marriot
Roundtable Knight
Roundtable Knight
Posts: 353
Joined: Thu Nov 20, 2008 3:02 am

Post by marriot » Tue Mar 15, 2011 3:09 am

Has anyone override systems positions in the last days?
Me not :cry:

Chelonia
Roundtable Knight
Roundtable Knight
Posts: 499
Joined: Mon Apr 30, 2007 3:37 pm

Post by Chelonia » Tue Mar 15, 2011 7:25 am

Nope, but steady -1.5% for March in the last 4, 5 days

LeviF
Roundtable Knight
Roundtable Knight
Posts: 1428
Joined: Mon Dec 22, 2003 12:24 pm
Location: Des Moines, IA
Contact:

Post by LeviF » Tue Mar 15, 2011 10:42 am

I got crushed overnight with my exposure to AUD. Nothing atypical though.

drm7
Roundtable Fellow
Roundtable Fellow
Posts: 96
Joined: Sun Apr 20, 2003 9:02 pm
Location: Richmond, VA

Post by drm7 » Tue Mar 15, 2011 1:26 pm

I would think that, for an automated system developer, I would be less worried about what's going on now, since my stops would take me out of "the wrong side of the market."

I would be more concerned about my system's ability to stay reasonably solvent after multiple, significant gaps through my stops (say, multiple simultaneous 5R losses.) Curtis Faith discussed one horror story with a Eurodollar position in his Way of the Turtle book.

I would be more tempted to override the system when things are going great, to "take chips off the table," so to speak. There's an interesting thread about everyone's reaction to the huge run-up in Cotton somewhere around here...

stopsareforwimps
Roundtable Knight
Roundtable Knight
Posts: 199
Joined: Sun Oct 10, 2010 1:47 am
Location: Melbourne Australia

Post by stopsareforwimps » Tue Mar 15, 2011 4:32 pm

Asamat wrote:However - coming back to the title of the thread - a black swan is not simply a big move or an unexpected move...
Taleb gives the example of his country of origin, Lebanon ("The Levant") where Christians and Moslems had lived together in peace for centuries. Then it suddenly all fell apart. His older relatives were expecting a return back to that "normality" but it did not happen. What seemed solid and secure disappeared suddenly and for good, for all intents and purposes.

One key characteristic of a black swan event is that you cannot analyze it statistically. This could be due to rarity, lack of data, lack of understanding of the underlying mechanisms, or just a level of complexity that makes it too hard to say much that is useful. Such a level of complexity is normal in human affairs. Why did the Roman Empire collapse? Why did it endure for so long? Did its collapse have anything to do with the fact it has just adopted Christianity as its official religion? Who knows? Many people have made up stories after the event.

A black swan is not just fat tails. Fat tails are amenable to analysis, to a certain extent. There are hundreds of nuclear reactors. Many of these are on geological fault lives. Many are old or are unproven designs. Many are poorly maintained or maintenance is not done honestly with true records being kept. Size 9.0 earthquakes have happened before. We know all this. So Japan is not a black swan.

When I do back testing, these kind of events are what limits my ability to increase heat. Currently I am looking at Ralph Vince's theory that you can use options to effectively remove these extreme events which allows you to run with more heat without risking being killed by one event.

I imagine Taleb's strategy of buying out of the money options is looking pretty good this week!

marriot
Roundtable Knight
Roundtable Knight
Posts: 353
Joined: Thu Nov 20, 2008 3:02 am

Post by marriot » Wed Mar 16, 2011 1:16 am

Well, in 1995, after Kobe Quake, Yen and bonds went up, Japanese Stocks went down 20 %.
Sure i will wait signals, because i don't know what to do, but exit from nikkey long positions on monday morning would have be better.

rabidric
Roundtable Knight
Roundtable Knight
Posts: 243
Joined: Mon Jan 09, 2006 7:45 am

Post by rabidric » Wed Mar 16, 2011 3:38 am

"..not amenable to statistical analysis..."

i sort of agree. but in a different way.

For me this does count as a black swan. It is largely unanticipatable in timing and size. The important thing though is that I have traded through this kind of thing many times now in my trading life, and not only have ways to "deal with it", but have also unlearned my root emotional responses, and am able to look for ways to profit. These profit opportunities do lie outside my "system" though- it is designed to trade the great majority of the passage of time in a mechanical/statistical way.


I guess what i am saying is that this event effectively overrides other people's trading plans(imagine all the long only mutual fund accumulation style strats in stocks over the last couple of years - their variance just got blown out the water) and short circuits directly to their animal fight or flight response. It may not be "statistically amenable", but it sure does become more predictable once you know what Mode the majority are operating in.

So then by some kind of heuristic/fuzzy logic based descision making , i find myself using fundamental based( the media have overblown the threat to long term economic activity in peoples minds), and intraday based technical analysis(normally poor expectation) to scalp some nice opportunities. Discretionarily.

i don't recommend this action to people who haven't done much profitable discretionary trading ever.

Post Reply