|
PureBytes Links
Trading Reference Links
|
Big,
I have problems with the whole idea of divergence. I think of it as
indicator failure instead.
While Wilder came up with the RSI, Walter Baeyens has written a book
on it recently that really goes into a lot more detail. Andrew
Cardwell is another who has also written a good bit about the RSI.
Basically, the RSI is a ROC that shows how many ups you have compared
to downs. So with a RSI(14) and an indicator value of 50 you have had
7 up days and 7 down days.
Finding 7 or 8 good stocks is really part of fundamental analysis.
Investools was great at developing the whole process. Also pay
attention to Super's posts. He can give you some very good ideas.
More Later,
Preston
--- In equismetastock@xxxxxxxxxxxxxxx, "Big Papa"
<denver69692002@xxx> wrote:
>
> Preston,
>
> Ok, I got the indicator in. And included the r1, r2, and r3, then
> smoothed the r2 by 3 and the r3 by 2.
>
> My initial reaction is that it seems to work well on 5-7 day
> increments,( I trade with EOD data), but doesn't catch the longer
> trends. I applied to the QQQQ, and lets say I bought at RSI<10 and
> sold at RSI>90. Many times over the past 5 months, it would have
> given false buy signals.
> But, when it does work, for instance, Oct 27th @ 32.54 to Nov 11th
@
> 36.93, it is good.
>
> When I review the meaning of the RSI, I am looking for a divergence
> between the market index making a new high, but the RSI failing to
> surpass a new high. That divergence would indicate a reversal.
Seems
> like I might need to track down Mr. Wilder's book?
>
> I think I read Steve's things posted here about the StochRSI. I did
> apply that to several stocks and really liked the way it looked. I
> applied standard error bands and did buys on the StochRSI and exits
> based on which bands it hit. I did most of the work on that in July
> and August. Needless to say, I got spooked by its performance from
> then until now. Autozone (AZO) was a great one for that system.
But,
> it seemed like Steve had it absolutley dialed in for the 7 or 8
> things he traded. I'm still jealous and frustrated I can't find my
7
> or 8 things.
>
> Big
>
>
> --- In equismetastock@xxxxxxxxxxxxxxx, pumrysh <no_reply@> wrote:
> >
> > Big,
> >
> > The RSI(2) is popular and seems that I also read a TASC article
on
> > the RSI(3). Once you break down the RSI formula you can see why
> the
> > short numbered lookbacks work so well and are becoming popular.
It
> is
> > really nothing that we haven't known for some time. Steve Karnish
> > loves and uses the StochRSI for very similar reasons.
> >
> > The RSI uses the ROC or C-ref(c,-10), sums the ups and downs,
> smooths
> > them using wilders smoothing and then scales to 100(maybe not in
> that
> > order).
> >
> > Using the RSI Raw Indicator plot 3 different RSIr's[RSIr(1),
(2),and
> > (3)] on a chart and you will see something rather unique.
> > Remember that the RSI Raw is not Wilder smoothed.
> >
> > The RSIr(1)will be either 0 or 100.
> > Now compare a RSIr(2) and RSIr(3).
> > Then smooth the RSIr(2) by 3 and the RSIr(3) by 2.
> > Now compare the results.
> >
> > Let me know what you think.
> >
> > Preston
> >
> >
> > ps. The RSI Raw Indicator is below if you need it:
> >
> > {Name: Rapid RSI
> > Formula:by Golson at Equis}
> > tp:=Input("Length",1,1000,14);{time periods in RSI calc}
> > plot:= C;
> > change:= ROC(plot,1,$);
> > Z:=Sum(If(change>0,change,0),tp);
> > Y:=Sum(If(change<0,Abs(change),0),tp);
> > Ytemp:=If(y=0,0.00001,y);
> > RS:=Z/Ytemp;
> > 100-(100/(1+RS)){end}
> >
> >
> >
> >
> >
> > --- In equismetastock@xxxxxxxxxxxxxxx, "Big Papa"
> > <denver69692002@> wrote:
> > >
> > > Preston,
> > >
> > > Thanks. I just got done reading Larry Connors latest
book "Short
> > > Term Trading Strategies That Work".
> > >
> > > He has a plan which he calls the VIX Stretch whereas he trades
> the
> > > SPY based on the following rules:
> > >
> > > SPY>200 MA
> > > VIX> 5% or more above the 10 day moving average for 3 or more
> days.
> > > Buy on close.
> > >
> > > Exit when SPY 2 day RSI > 65
> > >
> > > He offers up different scenarios that can be run including
> > different
> > > VIX % above the 10 MA plus the amount of days it is above the
> > moving
> > > average.
> > >
> > > All through the book, Mr. Connors preaches the value of the 2
> day
> > > RSI. Not sure if that is to lead folks to the TradingMarkets
> > website
> > > and his real cash cow of the subsciption to their stock rating
> > > system, but I suspect a little bit so for the unitiated.
> > >
> > > But overall, he does have 5-6 laid out strategies that could
> start
> > a
> > > beginner in the right direction, I think. One thing he does
well
> is
> > > back up the history of trades with the numbers. One thing he
> > doesn't
> > > do well is show how much money is lost on the 20-25% of trades
> that
> > > don't go well. And he is somewhat against stop losses, which
> scares
> > > me a bit from a risk and capital preservation standpoint.
> > > Disclosure, Big is not on the Connors payroll. :)
> > >
> > > Thanks for the help.
> > >
> > >
> > >
> > >
> > >
> > > --- In equismetastock@xxxxxxxxxxxxxxx, pumrysh <no_reply@>
wrote:
> > > >
> > > > Big,
> > > >
> > > > See if this helps.
> > > >
> > > > Preston
> > > >
> > > >
> > > > Buy:= C <=LLV(C,7);
> > > > {close is less than or equal to lowest low value of close
over
> 7
> > > days}
> > > >
> > > > Sell:= C >= HHV(C,7);
> > > > {close is greater than or equal to highest high value of
close
> > > over 7
> > > > days}
> > > >
> > > >
> > > >
> > > > A RSI(2) is of little value and I would suggest using the 3
> day
> > > > lookback period for it.
> > > > You can sum for 2 days like this:
> > > >
> > > > Sum(RSI(3),2)
> > > >
> > > > Using Cum() would add the values going all the way back to
the
> > > > beginning of the chart.
> > > >
> > > >
> > > >
> > > > For the VIX, I have:
> > > > VIX: The average implied volatility of 8 OEX options with a
30
> > day
> > > > expiration reported as an index.
> > > >
> > > > Volatility Spike: As defined by techies it is when the VIX
> rises
> > > > 35% .
> > > >
> > > > Saitta surmized that the volatility spike was useless in
> > > markets
> > > > that did not rise above 35%, therefore he standardized the
> > measure
> > > by
> > > > using a 15% rise of the 20 day moving average. His definition
> is:
> > > >
> > > > Saitta's Volatility Spike: The VIX minus 115% of its 20 day
> > moving
> > > > average. In metastock terms the formula would be:
> > > >
> > > > {Saitta Volatility Spike}
> > > > A:= P; {This must be dropped on VIX price plot}
> > > > B:= (mov(A,20,s) * 1.15)
> > > > X:= A ? B;
> > > > X;
> > > >
> > > > You would need to open two charts. One with a price plot of
> the
> > > S&P
> > > > 500 and another with the VIX. Click onto the VIX plot and
move
> it
> > > > onto the same chart as the S&P. Call the Saitta Volatility
> Spike
> > > > (SVS) indicator from the indicator list and drop into its own
> > > window
> > > > when the VIX plot is highlighted. Use a horizontal line at 0.
> > > >
> > > >
> > > > Finally,
> > > > ref(C,-1)>Open
> > > > is correct.
> > > >
> > > >
> > > > --- In equismetastock@xxxxxxxxxxxxxxx, "Big Papa"
> > > > <denver69692002@> wrote:
> > > > >
> > > > > Group,
> > > > >
> > > > > I need some quick checks on some coding.
> > > > >
> > > > > Buy: Close @ 7 day low, code - C<=(LLV(C,7))
> > > > >
> > > > > Sell: Close @ 7 day high, code - C>=(HHV(C,7))
> > > > >
> > > > > Correct?
> > > > >
> > > > > Next, I am looking for a cumulative value of RSI over a
> period
> > > of 3
> > > > > days.
> > > > >
> > > > > Cum(Sum(RSI(2),3)
> > > > >
> > > > > Correct?
> > > > >
> > > > > Next, I really need help on this one. Looking for a code if
> VIX
> > > >
> > > > 5%
> > > > > or more above the 10 MA for 3 or more days.
> > > > >
> > > > > Lastly, today open > yesterday's close
> > > > >
> > > > > ref(C,-1)>Open
> > > > >
> > > > > Thank you and Happy Holidays.
> > > > >
> > > >
> > >
> >
>
------------------------------------
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/equismetastock/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/equismetastock/join
(Yahoo! ID required)
<*> To change settings via email:
mailto:equismetastock-digest@xxxxxxxxxxxxxxx
mailto:equismetastock-fullfeatured@xxxxxxxxxxxxxxx
<*> To unsubscribe from this group, send an email to:
equismetastock-unsubscribe@xxxxxxxxxxxxxxx
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|