Adding to your portfolio

General discussions about futures.
Post Reply
damian
Roundtable Knight
Roundtable Knight
Posts: 815
Joined: Tue Apr 15, 2003 8:43 pm
Location: dusseldorf

Adding to your portfolio

Post by damian » Fri Jun 13, 2003 3:20 am

I have added gold to my long term trend portfolio.

New money to the account now permits one or two extra commodities. On my system, gold's performance is not fantastic, but the last year or so has produced better trades than some of the markets which I have actually been trading. Backtest metrics are improved by the addition and the equity curve benefits, although not by anything significant (to someone like me who likes to round numbers off). Gold would not not really merit inclusion on backtest results alone. It is an average-to-poor performer on my system showing largely breakeven results over simulation tests... but it does not detract form the overall portfolio either.

However, my main reason for looking at gold again: I feel that gold will have a different meaning as a commodity in the next 5 (or perhaps more) years than it has had in the last 10 or so years. I am left scratching my head as I suddenly started thinking in terms of fundamentals and being influenced by instinctive gut feel.

Does anyone else care to share what process they apply to selecting a new market to trade under an existsing system?

enjoy your weekend.

damian

(feel free to move. I didnt know where to put this as there is no portfolio theory or general realtime trading thread).
Last edited by damian on Fri Jun 13, 2003 8:42 am, edited 1 time in total.

damian
Roundtable Knight
Roundtable Knight
Posts: 815
Joined: Tue Apr 15, 2003 8:43 pm
Location: dusseldorf

Post by damian » Fri Jun 13, 2003 8:15 am

Hi Vince, thanks for you response.

Point noted on expectation, but give me any system and I will find a market on which it does not have a positive expectation. Besides, I measure expectation on a system like this over a portfolio rather than instrument. Essentially, I trade one 'market' that is made up of several products. The 'market' is the portfolio, the products are ED, CL, SF, KC etc. If I change the product mix, then I change the portfolio thus I have changed the 'market'.

I too question why my instinct influenced me. Ed probably has a point, however I do not need to hear it from him to make it a valid point. Besides, I once checked out what he had to say. In response to one readers email he said some thing like, but not exactly: "I suggest you learn how to speak English correctly. Dr Suess books are a good place to start". He then followed up with a picture of Green Eggs and Ham, or The Cat In The Hat, can't remember. Either way, I hit the back button then and continued to read theonion.com for some more interesting attempts at taking the p*ss.

Also, please excuse the spelling of Suess. It has been a while since I read one of the Dr's books. Ironically, if anyone needs to follow Ed's advice on literacy, I do
:wink:

cheers
damian

A follow up thought: why, if not for instinct and personally perceived perception of potential performance (welcome to the the five P's), did trend followers add EUR FX to their portfolios when it went live?

Vince
Roundtable Fellow
Roundtable Fellow
Posts: 61
Joined: Wed Jun 11, 2003 12:40 pm

Post by Vince » Fri Jun 13, 2003 10:33 am

Ed ........... taking the p*ss
Yes...Why is he such a pain in the a.s? It kills the message he is trying to get across. But he does have something to say, I believe.
why.......did trend followers add EUR FX
Because their systems have positive expectation. A system should have a lead-in period, where notional trading takes place, so that all of the decision making components of the system have a representative sample of data, upon which to make those decisions. Once, they have that, then the system does all the decisions for you.

My gut feeling, is that mechanical systems don't have a portfolio as such, because they are used to monitor everything in their space, which could mean every instrument traded in the world. The trade with the best probability of winning, based on the total systems, decision making process, is then automatically traded. Thats why they catch all the large moves whenever they occur. I guess you split the world of instruments into correlations, and position your money accordingly, so that, not all of your money goes into one highly correlated group.

damian
Roundtable Knight
Roundtable Knight
Posts: 815
Joined: Tue Apr 15, 2003 8:43 pm
Location: dusseldorf

Post by damian » Fri Jun 13, 2003 11:51 am

Hi Vince,

I asked:
>>why.......did trend followers add EUR FX
you, in part wrote: Because their systems have positive expectation.
I am not at all having a go at you (big smile to prove it: :D) but I think you may have missed my point. Their systems may well have had a positive expectation, but they didn't know if it would be so when trading EUR as it did not previously exist. Why then, without any empirical justification based on real history of teh real thing did they choose to trade EUR. My answer is gut instinct, the 5 P's. They had nothing else to base their decision on. Sure, they may have had great confidence that their systems had +E, but they also probably knew that their system had -E on the SP. They knew this for a fact. Perhaps they also knew that SF and DM traded well on their systems and followed that EUR would also trade well. This is 5P at work in portfolio composition, not mechanical mentality. Mechanical mentality would have had you sitting around for a few years to gather data and do a meaningful backtest.

d

Vince
Roundtable Fellow
Roundtable Fellow
Posts: 61
Joined: Wed Jun 11, 2003 12:40 pm

Post by Vince » Fri Jun 13, 2003 12:02 pm

Damian wrote:

"Why then, without any empirical justification based on real history of teh real thing did they choose to trade EUR. My answer is gut instinct, the 5 P's."


The question is "how much empirical justification do you require?" I would argue that you need a relatively small amount of data, upon which to make your trading decisions. Maybe as little as 10 trades will be sufficient, which might equate to one years worth of trading. Don't forget also, that the EURO was being traded long before the official launch, as the ECU. This was the currency as it was being traded in tandem against the "basket" of european currencies.[/quote]

Mark Johnson
Roundtable Knight
Roundtable Knight
Posts: 122
Joined: Thu Apr 17, 2003 9:49 am

Post by Mark Johnson » Fri Jun 13, 2003 2:06 pm

I added the euro currency (CSI symbol "CU") to my trendfollowing portfolio for a mundane reason: because I wanted to continue trading the Deutsche Mark in a highly liquid instrument after Germany converted to the Euro. DM had been very good to me in the past, both in backtesting and in live trading with real contracts, so I kept the ball rolling using CU. Haven't been disappointed, at least so far.

Vince
Roundtable Fellow
Roundtable Fellow
Posts: 61
Joined: Wed Jun 11, 2003 12:40 pm

Post by Vince » Fri Jun 13, 2003 2:14 pm

Good to see you're using CSI.

I don't think that this site is really geared towards "mundane" decision making processes. My instinct leads me to believe, that this site is dedicated to scientific and rigorous, testing and decision making, processes. No offence intended.

Just as a point of interest: "What Stop placement strategy are you using?"

When I'm trading the Euro (currencies in general), and I get a signal, I go in hard and tight (think of the missus when you're doing it - hehehe). Usually, I will whack up my unit size to 10% of my Net Equity per unit and then put on 5 units with 1/4 ATR spacing. I also ramp up, to max leverage, which with FXCM is 200:1. I then place one Stop to cover all units, a 1/2 ATR behind the first unit, and leave it there for the duration of the trade. I've survived on this "shock and awe" Stop strategy for years.
Last edited by Vince on Sat Jun 14, 2003 12:44 pm, edited 1 time in total.

damian
Roundtable Knight
Roundtable Knight
Posts: 815
Joined: Tue Apr 15, 2003 8:43 pm
Location: dusseldorf

Post by damian » Fri Jun 13, 2003 10:14 pm

Vince wrote:I don't think that this site is really geared towards "mundane" decision making processes. My instinct leads me to believe, that this site is dedicated to scientific and rigorous, testing and decision making, processes. No offence intended.
Yeah mark, it is about time you started researching like a mechanical trader. You should start by purchasing a good back testing engine. :wink: :wink: :wink: :lol:

Ted Annemann
Roundtable Knight
Roundtable Knight
Posts: 118
Joined: Tue Apr 15, 2003 7:44 pm
Location: Arizona

Long term trends in stock indexes

Post by Ted Annemann » Wed Sep 17, 2003 9:30 pm

For laughs I tried a long term trendfollowing system on US stock index futures. The Nasdaq, Russell, and S&P futures were net profitable over the last 3 years and also the last 6 years. Besides those, the NYFE and the Dow futures were net profitable over the last 3 years. I was a little surprised, I'd heard that trendfollowing the S&P was suicide.

Post Reply