Isn't it beautiful!

On a recent harbour cruise I just caught "Wild Oats" in the lens. It was great to see this magnificent yatch feature on the front page of the Sydney Morning Herald the next day!

Equity Curves

Equity curve of closed trades only to September 2005. Trades are on the X axis and include all contract notes, I haven't grouped similar trades, such as 3 sells on the same stock over a week. I also plotted a moving average line although I don't use it for anything.

DOM

I prefer not to post trades in play in case someone thinks its a recommendation. I picked this one up with a system that doesn't give a lot of signals. Hopefully it pans out ok. It had already moved substantially before I jumped on, with signals firing much earlier in other systems.

Take the trade, follow the system, don't think too much and sell when I get the signal. The trailing stop should keep me out of too much trouble.

Risk & Reward

Reward/risk ratio (simulated using TradeSim & AmiBroker) for the system shown in the WPL trade below. I took the biggest winner out. This is the only system I trade where the exit points stay fixed from the week that the trade is taken. Average positive RR was 2.58 and average negative RR was -0.70. I use 1% risk position sizing. This is a fairly new system so I am keeping a close eye on it. I designed it to trade the stocks in the ASX100. So far so good.

WPL Trade

Exited this one after profit target reached last Friday. It could easily go higher, but I'll stick to the system. $11 / share profit.

Steve

Monte Carlo with AB

I have been doing a bit of work on my spreadsheet to analyse Monte Carlo runs in AmiBroker. The screenshot shows my Graphs page.

I just do a dummy optimise in AB dump the portfolio summary into Excel and everything else is pretty much automatic. This one has 500 iterations.

I'm off to Portugal on Wednesday and probably won't post anything here for at least 3 weeks.

regards
Steve

Spirit of Ecstacy. Not my Roller though.

Posted by Picasa

Random entry

Try again - this chart is with a random entry.

Random entries and exits

The hypothesis is that entry criteria is irrelevant - that you could just buy at random and get just as good a result as an entry criteria that looks for some pattern.

I did some testing and found quite the opposite. Compound annual return for stocks using random entry and random exits are plotted below;



The results show that the random exit system outperformed the random entry system with the MA stop in terms of compound annual return. A random exit is really like a variable timed stop.

Quarterly returns

Above results are closed trade profits from trading mechanical systems on ASX stocks. I'm just practising on roughly $0.5 million before I get into serious trading :), although I keep spending the profits! Mechanical long term systems take time to build profits, I'm not looking at fast bucks, just steady consistent dollars.

Stevo

AWE

AWE also came up with a buy in one of my systems. I'm actually practising posting images on this site! They are a bit fuzzy at the moment. I was using Opera to post images and it didn't work so I'm back to IE.

Here's a more traditional view for the chartists.

Posted by Picasa

OXR for the more traditional chartists

It looks good - breakout on volume.

Lots of buy signals this weekend - More signals than cash. A preliminary look suggests OXR for starters using "G". I will have a look at the fundamentals of some of the candidates before making a decision. No sells for the week.

Posted by Picasa

WPL using "G" - this one I do hold. The system draws 2 exit points on entry - the red lower line is the loss target and the upper green line is the profit target. This one needs a weekly close above $34.44 to trigger an exit. An MA trailing exit point (orange line) completes the picture. It looks promising at the moment but who knows if it will hit it's profit target!

Posted by Picasa

ALN using a weekly system I developed using IO called Channel G. Admittedly a well chosen example - the exit in early 2004 was pure fluke! Also I did not make these trades as the system is only a few months old. I have a few trades running using this system - WPL, JHX, MAY and LHG from memory.

Posted by Picasa

Results from "weekly" system - Entry in May, exit in follwoing March. Universe of stocks was the ASX200 from 1998 to July 2005. System was developed on the ASX100 stocks, but appears to work on other stock universes. Drawdown is not brilliant at 15%, but for a lazy system it certainly shows promise. I used IO and AmiBroker to determine the system criteria as shown in the IO status "swarm" shown below.

Posted by Picasa

THis IO test shows an optimization on, of all things, buy and sell months. The buy month selected is May and the sell month is the following march. A 42 week ROC above 36 is a selection criteria, along with turnover. Exit trigger is either the month of March, or ROC below -23. Not suggesting that this is tradeable, it's just inspired by Dogs Of the Dow type strategies.

Posted by Picasa

MBL gave a "quick" 3 week trade for the super fund. Profit exits are pretty rare with this system. They trigger due to a rapid upward price move.

Posted by Picasa

JHX is a surprising trade, currently breaking out to new highs.

Posted by Picasa

Telstra - no signals on this one!

SFE trade using long term system

Posted by Picasa

Monte Carlo 10 positions. This one makes more than the 4 position test.

Posted by Picasa

Monte Carlo using only 4 positions - Percent Profit

Posted by Picasa

4 positions versus 10 positions

The 2 charts above are monte carlo simulations of a system that I have been working on. The 10 position position sizing gives better results in terms of drawdown than the the 4 positions only test.

See;

http://lightning.he.net/cgi-bin/suid/~reefcap/ultimatebb.cgi?ubb=get_topic;f=15;t=000236;p=1#000010

Posts for Reef

Seems like a good post to put up charts for Reef.