[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Help with EL for relative strength of stock vs sector



PureBytes Links

Trading Reference Links

Barry,

Here is a thought that I have tested and I have found valuable.  Compute the
relative performance as you have done but don't smooth it with your
"RatioLen."  Then use "Spread_B" as a long exit when it has failed to make a
new high in say 13 weeks.

I got this idea years ago from the newsletter "The Cabot Market Letter" and
I have found it to be a valid method for bailing out of a poor performing
stock or mutual fund.  Problem is when the whole market takes a dive RP
doesn't help that much.  I built a trading system that included an entry and
exit.  If I remember the system it worked well with weekly mutual funds.  I
was going to combine this (using Fuzzy Logic) with my version of Nicholas
Darvis box method  for my "killer" mutual fund trading system.  Problem is
that I never have enough time for trading system development.  Anyway, I'll
try to find which computer I have it on and I'll post the EL code on this
newsgroup.

Marlowe

----- Original Message -----
From: Barry Silberman <barry@xxxxxxxxxxxxxxxxxxxxx>
To: <omega-list@xxxxxxxxxx>
Sent: Friday, March 03, 2000 10:25 PM
Subject: Help with EL for relative strength of stock vs sector


> In trying to backtest  improvements to entries for EOD trading with
equities
> I came up with the following code to measure relative strength of an
> individual stock (data1) as compared with a sector (Data2).  Although the
> code seems to work, it suprisingly improves the entries only marginally.
>
> Can someone tell me if there is a flaw in the reasoning or a  better way
to
> approach this concept of relative strength.
>
> The code I'm using is as follows:
>
> {Sector Strength Filter
> Data1 = individual stock
> Data2 = sector  }
>
> INPUTS: RatioLen(5);
> Variable: spread_B(0), DataOne(0), DataTwo(0);
>
> DataOne = xaverage(close data1, RatioLen);
> DataTwo = xaverage(close data2, RatioLen);
>
> If DataTwo <> 0 then
> Spread_B = DataOne / DataTwo;
>
> Condition1 = spread_B - spread_B[1] > 0 ;   {For Long entry}
> Condition2 = spread_B - spread_B[1] < 0;   {for short entry}
>
>
> Thank you.
>
> Barry
>