Page 5 of 5

Posted: Thu Apr 21, 2011 9:56 am
by Chuck B
Just to clarify...

The exit you used is initially 7ATR from the entry price.

Ongoing I assume you meant that this 7ATR exit was updated (referenced off the close (or high/low) each day) and only moved in a favorable direction. Hence this is the unconditional exit -- a trade will ALWAYS exit if this is hit.

Is the 7ATR exit on a touched basis or a closing below basis?

Now the conditional exit is the mov avg cross-over such that it the trailing 7ATR exit hasn't been hit, but a SMA/MMA cross occurs on the close, the position is exited at the next market open?

You didn't mention the scale out algorithm. I assumed it was based on open position risk as a percentage of portfolio equity, but that was my bias reading your post, lol. I suppose you might have price-based rules for that too but neither is specified.

Thanks for the update...just curious questions here so that I understand what you have posted.

Cheers,
Chuck

Posted: Thu Apr 21, 2011 10:21 am
by AFJ Garner
The exit you used is initially 7ATR from the entry price.
Yes
Ongoing I assume you meant that this 7ATR exit was updated (referenced off the close (or high/low) each day) and only moved in a favorable direction. Hence this is the unconditional exit -- a trade will ALWAYS exit if this is hit.
No, the stop was left unchanged as per the built in TB TMA system (IE non-trailing). This is something I have changed for my own systems.
Is the 7ATR exit on a touched basis or a closing below basis?
On a closing below basis.
On a closing below basis.
Now the conditional exit is the mov avg cross-over such that it the trailing 7ATR exit hasn't been hit, but a SMA/MMA cross occurs on the close, the position is exited at the next market open?
Correct
You didn't mention the scale out algorithm. I assumed it was based on open position risk as a percentage of portfolio equity, but that was my bias reading your post, lol. I suppose you might have price-based rules for that too but neither is specified.
50% of position sold each time the price moves 10 ATR in a favourable direction.. Once the target is hit, the next target is then calculated with reference to the then current ATR.

I make no special claims for this system - it is pretty well an "out of the box TB built in system". It was simply a bog standard test of a bog standard system using an almost random selection of stock indices to make a point.

The real point being: "go it alone; do not buy mutual funds; do not employ professionals; buy and hold index trackers or employ a simple system or both". I do not have great respect or admiration for the established fund management industry.

best
A

Posted: Thu Apr 21, 2011 10:46 am
by nodoodahs
AFJ Garner wrote:The real point being: "go it alone; do not buy mutual funds; do not employ professionals; buy and hold index trackers or employ a simple system or both". I do not have great respect or admiration for the established fund management industry.
I would change to "do not buy actively-managed discretionary mutual funds, limit funds to either index funds or those employing easily-understood systematic trading methods without management discretion applied;"

I would change to "buy and periodically rebalance diverse multi-asset index trackers ..." but keep in mind that "rebalancing" is a simple system relying upon mean reversion of performance across asset types.

Totally ditto the last sentence there.

Posted: Sun Aug 21, 2011 9:52 am
by AFJ Garner
What was that word I was grasping for? Ah yes, “Hubrisâ€

Posted: Sun Aug 21, 2011 9:56 am
by AFJ Garner
Here is the portfolio:
AEX
ALS
BFX
DJ
FCH
FDX
FFI
HCE
HSI
IFS
KLI
KOS
MFX
NK
NQ
O30
SP
SSG
STW
SWI
SXE
SXF
YAP2

Posted: Sun Aug 21, 2011 10:22 am
by AFJ Garner
By way of contrast here is the same system (same in every detail and parameter) run on a mixed portfolio of over 100 futures. Current theoretical positions comprise bonds, interest rates, gold and a few currencies.

Posted: Mon Aug 22, 2011 8:56 am
by drm7
[quote="AFJ Garner"]What was that word I was grasping for? Ah yes, “Hubrisâ€

Posted: Mon Aug 22, 2011 12:01 pm
by AFJ Garner
drm7 wrote: That being said, I have a few follow-up questions:

1) What are the results of a system that allows shorting?
2) What does the buy and hold equity curve look like? Did your LTTF equity curve stay above it?
3) What about removing profit targets/scaling out?
The following results are based on the identical system and parameters first set out with the identical stocks only portfolio (except as mentioned).

(1) Shorting greatly improves the CAGR but not the recent severe DD - see first screen shot below ("add shorts") which goes long and short. But remember the idea of the article was to give an alternative to retail buy and hold. For most of those people shorting is probably too complex and too scary.

(2) Yes, the LTTF results remain well above the B&H despite the recent severe DD. See third screenshot ("Original").

(3) Removing profit taking helped the long only test (see screen shot 2 "Long Only no profit take") but I'm firmly in the pro profit takers' camp.

Posted: Sat Feb 25, 2012 2:29 pm
by Chelonia
Invert trades :-)