|
Here the report long only, $11 per trade, portfolio
backtest from 1995 YTD on NDX 100 component stocks using EOD data I
could be missing something but I see negative CAR, very high drawdown
with lots of trades. If you look at the charts you'll see 477 trades
losing 0-5% that were stopped out. To me that's a not a
good entry on this family of
stocks because all of your gains are attrited by
the losing trades that are stopped out. I don't see much to reccomend
this strategy as it is
and possibly it was not meant for end of day
trading.
Another problem is that the NDX component stocks have
changed since 1995, and running a back test like this
starting with 2005 NDX 100 gives it what's called a
survivorship bias...the survivors are now in the NDX 100 while
the non performers were dropped.
I don't have futures data.
I don't know if this helps
JOE
----- Original Message -----
Sent: Wednesday, August 17, 2005 8:50
AM
Subject: [amibroker] Re: Can somebody
test this simple system and post results ?
Portfolio (for stocks)- only long with commissions of
$11. For at
least 5-10 yrs.
For futures (if you have it) - both
long and short with slippage and
commissions of, maybe, $75. 5-10
Yrs.
Thanks
--- In amibroker@xxxxxxxxxxxxxxx, "Joe
Landry" <jelandry@x...> wrote:
>
Hello - Dickie, How do you want it run? Individual stocks backtest
or portfolio? Range?
>
> I ran it as a portfolio
backtest against NDX100 component stocks
and didn't get anything positive
for CAR.
>
> If I run it as individual backtest for this year
(YTD) I get a CAR
distribution, with the top 20 starting at a CAR
of
> 47% looking like this to a car of -43%. Almost looks like a
bell
shaped curve distribution.
>
> Looks like it
needs 'something'.
>
> Hope this helps
> JOE
>
>
>
> Ticker Net
Profit Net % Profit Exposure % CAR RAR Max. Trade
Drawdown
> GENZ 2690.94 26.91 7.01 47.19
673.59 -399.94
> PTEN 2362.2 23.62
3.18 41.06 1289.24 -268.01
> EBAY
2282.21 22.82 8.28 39.58 478.02 -636.19
> ERTS 2215.96 22.16 7.64 38.36
501.9 -284.99
> QCOM 2126.12 21.26
7.64 36.71 480.35 -597.19
> TLAB
2101.32 21.01 2.55 36.26 1423.27 -112.22
> YHOO 1985.41 19.85 6.37 34.15
536.16 -416.49
> PAYX 1788.71 17.89
4.46 30.6 686.26 -171.93
> AMZN
1633.94 16.34 2.55 27.83 1092.22 -224.55
> PETM 1475.05 14.75 5.1 25.01
490.77 -178.67
> ADBE 1445.25 14.45
3.82 24.48 640.59 -220.1
> CDWC
1439.62 14.4 6.37 24.38 382.79 -459.35
> JDSU 1420.79 14.21 3.82 24.05
629.3 -880.89
> BRCM 1380.04 13.8
4.46 23.33 523.32 -396.98
> LAMR
1368.83 13.69 7.01 23.14 330.21 -387.15
> VRSN 1351.03 13.51 3.18 22.82
716.64 -310.13
> DISH 1346.31 13.46
5.73 22.74 396.69 -67.49
> CTXS
1301.72 13.02 2.55 21.96 861.88 -343.07
> SEBL 1257.06 12.57 3.18 21.18
664.98 -412.26
>
>
> ----- Original Message
-----
> From: Dickie Paria
> To:
amibroker@xxxxxxxxxxxxxxx
> Sent: Wednesday, August 17,
2005 7:55 AM
> Subject: [amibroker] Can somebody test this
simple system and
post results ?
>
>
>
This is a slight modification to a simple system from an article
in
> April 2005 'Active Trader' mag (article by Xavier
Raj). It looks
> for 'extreme' price bars i.e., it
goes long the next day if price
> today opens at the bottom
10% of the price bar and closes at the
top
> 10% of the
bar. The trade stays in the market for 24 hrs i.e.,
it
> sells on the open of the following day. Going short
is the exact
> opposite.
> Can somebody
test it for me on the NDX100 stocks or, perhaps,
Dow30
?
> [I don't carry alot of stocks on my flash drive]
Thanks
>
> Range = (H-L);
>
> //Open is in the lowest 10% of the price bar and Close
> // is in the highest 10% of the price
bar
> PriceUp = Close > (High - (Range/10)) AND Open <
(Low +
(Range/10));
>
> //Open is in the highest
10% of the price bar and
> // Close is in lowest 10% of the
price bar
> PriceDown = Close < (Low + (Range/10)) AND
Open > (High -
(Range/10));
>
> //Trade on
BuyPrice and ShortPrice with one bar delay
> SetTradeDelays
(1,1,1,1);
>
> BuyPrice = ShortPrice = CoverPrice =
SellPrice = Open;
>
> Buy = IIf (PriceUp, BuyPrice,
0);
> Sell = 0;
>
> Short = IIf
(PriceDown, ShortPrice, 0);
> Cover = 0;
>
> ApplyStop (stopTypeNBar, 1, 1,0);
>
>
>
>
> Please note that this group is for
discussion between users only.
>
> To get support
from AmiBroker please send an e-mail directly to
> SUPPORT
{at} amibroker.com
>
> For other support material
please check also:
> http://www.amibroker.com/support.html
>
>
>
>
>
>
--------------------------------------------------------------------
----------
>
YAHOO! GROUPS LINKS
>
> a.. Visit
your group "amibroker" on the web.
>
> b.. To unsubscribe from this group,
send an email to:
>
amibroker-unsubscribe@xxxxxxxxxxxxxxx
>
> c.. Your use of Yahoo! Groups is
subject to the Yahoo! Terms
of Service.
>
>
>
--------------------------------------------------------------------
----------
Please note that this group is for discussion between users only.
To get support from AmiBroker please send an e-mail directly to
SUPPORT {at} amibroker.com
For other support material please check also:
http://www.amibroker.com/support.html
YAHOO! GROUPS LINKS
|