Explorations are not very good at picking stocks to buy, no matter
what
the constraint equations are or how many backtests have been
run.
Discretionary judgment is always necessary to pick among the
candidates.
Explorations should probably be called preliminary
screenings. When I
first started trading many years ago, I tried the
concept of
explorations as preliminary screens and tested the results. I
found
that using screens based on fundamental analysis and
quantitative
analysis produced a far superior pool of trading candidates,
so now I
only trade equities from Value Line's T100 list or IBD 100. The
only
exception is the ETFs I use for asset allocation strategies.
Somebody will probably post that they need mechanical systems to
take
the emotion out of trading. I've written several articles on why
the
only thing mechanical systems will take out is money from the
user's
wallet, so I'm not going to repeat all of that here.
As far
as shorting goes, it's a much more difficult method of trading.
There are
formulas for X% off of high. They're easy so someone else
can supply it to
you, but you might want to consider forgoing shorting
until you are
consistent and really good at long side trading. You'll
have a better
chance of success trading the long side, but only in
uptrends. Trying to
trade through sideways markets and downtrends is
not a very productive
idea, but everyone seems to need to reinvent the
wheel themselves or give a
lot of money to the so called "gurus"
before they realize the "gurus" don't
trade for a living. As I've said
before, they sell products. Almost all of
the guru's I know
personally, and I know several, use an asset allocation
investment
strategy for the majority of their money. Sometimes they keep
just a
little in the their trading account so they can say they trade,
and
then some don't trade at all. They never publish any prior long
term
trading results or backtests reports using their methods.
In
my experience, I've found the guys on the motley fools mechanical
investing
board make more money more consistently than the guys who
are on the TA
boards. I've compared the long term trading records from
as many
practitioners of both methods as I've been able to find.
Their version
of a mechanical system is based on FA and QA and buying
and holding for
specific periods of time. That kind of mechanical
system works fairly well.
That's one reason why I use a hybrid with
the VL stocks and IBD stocks.
Anyone who wants to spend some time looking can find the same data
and
draw their own conclusions from it.
Anyway, whatever your cup
of tea is drink up and have fun! Tea can get
expensive so drink sparingly
at first.
Super
PS Don't forget to test everything thoroughly
for long and short
strategies. Don't trust any set of equations that
haven't been tested,
no matter where they came from. And remember, you're
always going to
do worse than the tests results suggest so make sure the
Monte Carlo
testing covers a lot of possible situations and there is a wide
moat
of positive possibilities.
--- In equismetastock@yahoogroups.com,
"Big Papa" <denver69692002@...>
wrote:
>
> Thanks,
that helped out a bunch. Both of the suggestions seem to
> increase the
amount of stocks that pop up on the screen. Not sure if
> that is good
or bad.
>
> Still looking for some help on the "if it goes down X
%, then short,
> part". Is there a way to even program such a
thing?
>
>
> --- In equismetastock@yahoogroups.com,
superfragalist <no_reply@>
> wrote:
> >
> >
Oops,
> >
> >
H1:=If(H>=HHV(H,60),1,0);
> >
> > should be
> >
> > H1:=If(H=HHV(H,60),1,0);
> >
> > No big deal, the extraneous > does nothing.
> >
> > Super
> >
> >
> > --- In equismetastock@yahoogroups.com,
superfragalist <no_reply@>
> > wrote:
> > >
>
> > Big,
> > >
> > > Here's a formula I've
tested and written about for Roy's
> newsletter. I
> > >
think it will give you what you're looking for. You can tweak the
> >
> parameters a bit but I don't think optimization really brings
>
much to
> > > this party.
> > >
> > >
V1:=Mov(V,90,S);
> > > V2:=Mov(V,5,S);
> > >
VBuy:=If(V2>V1*1.5,1,0);
> > >
> > >
> > > H1:=If(H>=HHV(H,60),1,0);
> > >
> > > H1>0 and VBuy>0
> > >
> > >
Super
> > >
> > >
> > >
> >
>
> > >
> > >
> > > --- In equismetastock@yahoogroups.com,
pumrysh <no_reply@> wrote:
> > > >
> > > >
Big Papa,
> > > >
> > > > Had some time to
review and play with your exploration formula.
> > > >
>
> > > Here's what I'm seeing.
> > > >
> >
> > There are two important factors being explored here.
> >
> >
> > > > First, the HHV is acting as a trailing
stop/indicator. Using
> the
> > > > lookback period of
260 will keep you out of consolidation and
> whipsaw
> > >
> markets but has the drawback of getting you into a trade late.
>
> > > Realize that 260 periods would encompass the highs of over a
> years
> > > > period of time. Several changes that
you could consider would
> be to
> > > > shorten the
lookbacks or use either a moving average or ATR
> based
> >
> > indicator here.
> > > >
> > > >
Second, volume is a fickled lady. I am somewhat skeptical that
> it is
> > > > ever reported accurately. We also seldom know for sure
whether
> its
> > > > buying volume or selling volume.
In your case the exploration
> is
> > > > turning up
more selling volume than buying. This is probably
> due to
>
> > > the restriction that the HHV is putting on it. By using the
> single
> > > > day volume in comparison to its moving
average you are seeing
> more
> > > > spikes which can
also be throwing off your exploration. You
> may want
> >
> > to consider using short term moving averages of volume
>
instead.
> > > > Something like:
> > > >
Mov(V,5,S) > Mov(Mov(V,5,S),5,S)
> > > >
> >
> >
> > > > Hope this helps,
> > > >
> > > > Preston
> > > >
> > >
>
> > > >
> > > >
> > > >
--- In equismetastock@yahoogroups.com,
pumrysh <no_reply@>
> wrote:
> > > > >
>
> > > > Big,
> > > > >
> > > >
> PDO is one of those that you would like to find especially
> if
you
> > > > > could find it in April and sell it in May and
before it
> tanked on
> > > > the
> > >
> > 22nd.
> > > > >
> > > > > What
I noticed was that it fell into a trend, albeit a short
> one.
>
> > > >
> > > > > Question is how to do that
and not to catch the ones that
> are
> > > > highly
> > > > > volatile and tank the next day.
> > >
> >
> > > > > Several things I would play with. The
HHV has a 260 period
> > > > lookback.
> > > >
> I would probably shorten that. I would probably up the
> initial
> > > > stock
> > > > > price as well. $2
stocks are usualy being played/manipulated
> by the
> > >
> > hucksters, so I would probably bump that number to $10.
>
> > > >
> > > > > A good trend filter would
also be nice to include. Several
> come to
> > > > >
mind. Obviously the idea is going to be to jump on the trade
> early
> > > > > rather than late. Let me play with a few and see
what looks
> good.
> > > > >
> > > >
> Finally, you are going to need a very good trailing stop. I
>
think
> > > > > Andy just mentioned a good one but there
are others.
> > > > >
> > > > > Will get
back to you with more in a bit.
> > > > >
> > >
> >
> > > > > Preston
> > > > >
> > > > >
> > > > >
> > >
> >
> > > > >
> > > > >
>
> > > > --- In equismetastock@yahoogroups.com,
"Big Papa"
> > > > > <denver69692002@>
wrote:
> > > > > >
> > > > > > What
I have found is that the stock will behave wildly one
> way or
>
> > > > > the other, right after it is found in the
exploration.
> > > > > >
> > > > >
> For instance, it caught PDO around Mid May around $14.85
> before
> > > > it
> > > > > > took off to 15.74
the next day, $22.85 on day 3, etc.
> > > > > >
>
> > > > > It also caught UFPT around the same time, it crossed
my
> > > > magical "X"
> > > > > >
percentage, and was down 35% 20 trading days later.
> > > >
> >
> > > > > > During the first week in June, it
caught XCO at $26.96,
> which was
> > > > > at
> > > > > > $34.31 10 days later, and GEOI at $27.13,
down 23% 10 days
> later.
> > > > > >
>
> > > > > Thoughts?
> > > > > >
>
> > > > >
> > > > > > --- In equismetastock@yahoogroups.com,
pumrysh <no_reply@>
> wrote:
> > > > > >
>
> > > > > > > Big Papa,
> > > >
> > >
> > > > > > > I believe somethings
askew with this exploration.
> > > > > > > C>0
should be C>Open not zero.
> > > > > > > Next when
I plotted the code as an indicator on a price
> chart
> >
> > of
> > > > > > > several stocks what I saw
was selling opportunities not
> buying.
> > > > >
> >
> > > > > > > Preston
> > >
> > > >
> > > > > > >
> > >
> > > >
> > > > > > > --- In equismetastock@yahoogroups.com,
"Big Papa"
> > > > > > > <denver69692002@>
wrote:
> > > > > > > >
> > > > >
> > > I have an explorer. End of Day type set up.
> > >
> > > > > C>=2
> > > > > > > >
AND
> > > > > > > > C>0
> > > >
> > > > AND
> > > > > > > >
C>=HHV(H,260)-HHV(H,260)*0.10
> > > > > >
> > AND
> > > > > > > >
C>Ref(H,-1)
> > > > > > > > AND
> >
> > > > > > Mov(V,30,S)>1000
> > > >
> > > > AND
> > > > > > > >
V>=Mov(V,30,S)*2
> > > > > > > >
>
> > > > > > > I have been paper trading this for a month
or so, on
> next
> > > > day,
> > > >
> > > > 3,5,10, and 20 day results.
> > > > >
> > >
> > > > > > > > My buy formula will
buy all stock that pops up on the
> screen.
> > > > I
> > > > > > buy
> > > > > > >
> at the opening of the next day.
> > > > > > >
>
> > > > > > > > My sell formula is a little
bit trickier.
> > > > > > > > If the stock
declines more than x% from the close of
> the day
> > >
> > > > > previous to when I buy, I want to sell, and move to a
> short
> > > > > > > position
> >
> > > > > > right away. The equities that don't stop out,
I'll let
> run.
> > > > > > > > I need help
with that formula, both for the sell of
> the
> > > >
initial
> > > > > > > buy,
> > > >
> > > > plus the automatic buy of the short position.
> >
> > > > > >
> > > > > > > > Any
help for the exit would be nice too.
> > > > > > >
>
> > > > > > > > Thanks,
> > >
> > > > > Big Papa.
> > > > > > >
>
> > > > > > >
> > > > >
>
> > > > >
> > > >
> >
>
>
>
>