Thanks Howard for your comments.
The criteria is: use the ETF family as used by Connors,use the TPS code in this
thread, use daily data, then run a backtest with scaling in, run it without
scaling in (taking a full position at the first buy condition). I have tried
scaling in with other strategies too, and in general, my observation is results
are better without using scaling consistent with Howard's experience.
Regards
Larry
--- In [email protected], "longt3rm" <longt...@...> wrote:
>
> "...did not...worked better than just buying a full position"
>
> What is the criteria?
>
> --- In [email protected], "onelkm" <LKMCD1@> wrote:
> >
> >
> >
> > I also read Connor's book and found many of his strategies worked, probably
> > because they were simple. I did not find that scaling in, as is being
> > discussed in this thread, worked better than just buying a full position.
> > Anyone else come up with the same conclusions?
> > Larry
> >
> > --- In [email protected], "Richard" <richpach2@> wrote:
> > >
> > > Bill and Mike,
> > >
> > > Yes, this is the book I was referring to. I learned not to trust
> > > anything, so I test anything I read and Larry Connor's systems do match
> > > the results he publishes in his books.
> > >
> > > Mike, thanks for the hint for the "4 out of 5". I thought about using
> > > SUM() function for this but, that was after seeing how you applied it in
> > > the scale-in code.
> > >
> > > Kind Regards
> > > Richard
> > >
> > >
> > > --- In [email protected], "Mike" <sfclimbers@> wrote:
> > > >
> > > > Charting the interesting bars makes it a little easier to see
> > > >
> > > > Colors = IIF(InterestingBar, colorRed, colorDarkGrey);
> > > > Plot(Close, "Price", Colors, styleBar);
> > > > PlotShapes(Trigger * shapeUpArrow, colorGreen);
> > > >
> > > > Mike
> > > >
> > > > --- In [email protected], "Mike" <sfclimbers@> wrote:
> > > > >
> > > > > Hi,
> > > > >
> > > > > Which book/course are you guys referring to?
> > > > >
> > > > > As for "programming with ease", the best you can do is read lots and
> > > > > lots of coding examples. The rest you're born with. For better or for
> > > > > worse, your brain just works that way or it doesn't.
> > > > >
> > > > > As for 4 out of 5, no loop necessary. Loops should be treated as a
> > > > > last resort. Instead, try something like the following:
> > > > >
> > > > > InterestingBar = Low < Ref(Low, -1) AND High < Ref(High, -1);
> > > > > Trigger = Sum(InterestingBar, 5) >= 4;
> > > > >
> > > > > Plot(Close, "Price", colorDarkGrey, styleBar);
> > > > > PlotShapes(Trigger * shapeUpArrow, colorGreen);
> > > > >
> > > > > Mike
> > > > >
> > > > > --- In [email protected], "Richard" <richpach2@> wrote:
> > > > > >
> > > > > > Bill and Mike,
> > > > > >
> > > > > > Thanks for sharing your coding skills. Coincidently I was looking
> > > > > > at the same system.
> > > > > > If this is TPS then, it is one down and six to go. I have to say
> > > > > > that this one was the hardest to code from the book. I have two
> > > > > > questions, if I may.
> > > > > > How do one gain the skills necessary to program in AFL with such
> > > > > > (perceived) ease?
> > > > > > Have you coded MDU system? How did you solve the AFL definition for
> > > > > > "4 out of last 5 bars making lower low and lower high"?
> > > > > > Do I need a loop for this?
> > > > > >
> > > > > > Kind Regards
> > > > > > Richard
> > > > > >
> > > > > >
> > > > > > --- In [email protected], "longt3rm" <longt3rm@> wrote:
> > > > > > >
> > > > > > > Mike,
> > > > > > >
> > > > > > > Fantastic, it works!
> > > > > > >
> > > > > > > Thank you for your help!
> > > > > > >
> > > > > > > Bill
> > > > > > >
> > > > > > > --- In [email protected], "Mike" <sfclimbers@> wrote:
> > > > > > > >
> > > > > > > > Bill,
> > > > > > > >
> > > > > > > > You would need to calculate the trigger first (i.e. ExRem of
> > > > > > > > position state), then use ValueWhen.
> > > > > > > >
> > > > > > > > e.g. Something like the following:
> > > > > > > >
> > > > > > > > thirdTrigger = ExRem(inThirdPos, Sell);
> > > > > > > > bull400 = Close < ValueWhen(thirdTrigger, Close) AND inThirdPos
> > > > > > > > AND Ref(inThirdPos, -1);
> > > > > > > >
> > > > > > > > The key being that Flip will give all 1's for the duration that
> > > > > > > > we are in the position, whereas ExRem will remove redundant 1's
> > > > > > > > leaving only those bars where the condition was first met.
> > > > > > > > Thus, ValueWhen gives the Close of the most recent bar meeting
> > > > > > > > that condition.
> > > > > > > >
> > > > > > > > Mike
> > > > > > > >
> > > > > > > > --- In [email protected], "longt3rm" <longt3rm@> wrote:
> > > > > > > > >
> > > > > > > > > Mike,
> > > > > > > > >
> > > > > > > > > Thank you! That helps significantly!!
> > > > > > > > >
> > > > > > > > > The one item I'm not able to figure out is how to track the
> > > > > > > > > last entry price so we can check to see if the current close
> > > > > > > > > is lower than our last entry.
> > > > > > > > >
> > > > > > > > > I believe (I could be mistaken) the sample you kindly
> > > > > > > > > provided is checking to see if today's close is less than
> > > > > > > > > yesterday's close. I would like to compare today's close to
> > > > > > > > > our last entry price.
> > > > > > > > >
> > > > > > > > > Thanks again for helping so much. I am slow to grasp AFL.
> > > > > > > > >
> > > > > > > > > Sincerely,
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > Bill
> > > > > > > > >
> > > > > > > > > --- In [email protected], "Mike" <sfclimbers@> wrote:
> > > > > > > > > >
> > > > > > > > > > Hi,
> > > > > > > > > >
> > > > > > > > > > I posted some code to a related question here:
> > > > > > > > > > http://finance.groups.yahoo.com/group/amibroker/message/146956
> > > > > > > > > >
> > > > > > > > > > Short answers:
> > > > > > > > > > 1. Use Flip function to capture current position "state".
> > > > > > > > > > 2. Use state captured above AND'ed with next criteria.
> > > > > > > > > >
> > > > > > > > > > See if that helps.
> > > > > > > > > >
> > > > > > > > > > Mike
> > > > > > > > > >
> > > > > > > > > > --- In [email protected], "longt3rm" <longt3rm@>
> > > > > > > > > > wrote:
> > > > > > > > > > >
> > > > > > > > > > > Hello,
> > > > > > > > > > >
> > > > > > > > > > > Goal Description:
> > > > > > > > > > > 1) Initial $500 position opened when RSI(2) < 25
> > > > > > > > > > > 2) Add $1000 to position when entry price for #1 is less
> > > > > > > > > > > than current close. (regardless of value for RSI(2))
> > > > > > > > > > > 3) Add $1500 to position when entry price for #2 is less
> > > > > > > > > > > than current close. (regardless of value for RSI(2))
> > > > > > > > > > > 4) If all three positions are entered, we would have a
> > > > > > > > > > > total position of $3,000
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > Two questions / problems:
> > > > > > > > > > > 1) What is the most efficient way to determine if we have
> > > > > > > > > > > a position and if we do, is it the first ($500) or second
> > > > > > > > > > > ($500+$1000=$1500)?
> > > > > > > > > > > 2) If the initial buy is "Buy = RSI(2)", how do we tell
> > > > > > > > > > > AmiBroker that RSI(2) is not longer needed, we just need
> > > > > > > > > > > current close > last position entry price.
> > > > > > > > > > >
> > > > > > > > > > > Thanks for any suggestions.
> > > > > > > > > > >
> > > > > > > > > > > Sincerely,
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > Bill
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>
