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

RE: [amibroker] Re: QP2 - actual and backadjusted prices



PureBytes Links

Trading Reference Links




hhmm... I don't know 
where the confusion is coming from, but I'll try again.
<FONT face=Arial 
size=2> 
Every record in my 
database constains O,H,L,C,V and OI.
<FONT face=Arial 
size=2> 
O,H,L,C are 
backadjusted prices, similar to what you would see in VectorVest, TC2000 and 
just about every other data source.
<FONT face=Arial 
size=2> 
My V field contains 
actual volume.  I have no need for backadjusted volume.
<FONT face=Arial 
size=2> 
My OI field contains 
the actual close on the day.
<FONT face=Arial 
size=2> 
So... something like 
RSI in AB will use backadjusted prices, all smoothed out, compensating for 
splits, dividends, etc.
<FONT face=Arial 
size=2> 
My filter will use 
the OI field.    So a buy statement could 
say:
<FONT face=Arial 
size=2> 
Buy = OI > 100 
and Volume > 300000 and RSI() < 20;    // OI contains 
actual close, Volume contains actual volume and RSI is going to use the 
value in the close column.
<FONT face=Arial 
size=2> 
If I want to use 
average actual turnover in my filter, I could say:
<FONT face=Arial 
size=2> 
Buy = OI > 100 
and MA(Volume * OI / 100) > 400000 and RSI() < 20;   
//   OI = 100 = $1.00
<FONT face=Arial 
size=2> 
The "picking" and 
the "backtesting" are all done in one line of code.
<FONT face=Arial 
size=2> 
How hard can it 
be?
<BLOCKQUOTE 
>
  -----Original 
  Message-----From: Al Venosa 
  [mailto:advenosa@xxxxxxxxxxxx]Sent: Sunday, June 22, 2003 8:19 
  AMTo: amibroker@xxxxxxxxxxxxxxxSubject: Re: [amibroker] 
  Re: QP2 - actual and backadjusted prices
  But Chuck, I thought you said you only use pre-adjusted shares for 
  filtering, not for backtesting. For backtesting, you use split shares as 
  downloaded to get all your backtest performance statistics. I get the 
  impression from this post that you are also testing with pre-adjusted shares. 
  Once you pick the stocks that were between $1 and $10 10 years ago, do you 
  backtest them with that price or do you backtest them with post-split prices? 
   Am I misquoting you? 
   
  AV
   
   
  <BLOCKQUOTE 
  >
    ----- Original Message ----- 
    <DIV 
    >From: 
    <A title=chuck_rademacher@xxxxxxxxxx 
    href="">Chuck Rademacher 
    To: <A title=amibroker@xxxxxxxxxxxxxxx 
    href="">amibroker@xxxxxxxxxxxxxxx 
    Sent: Saturday, June 21, 2003 11:25 
    PM
    Subject: RE: [amibroker] Re: QP2 - 
    actual and backadjusted prices
    
    <FONT face=Arial color=#0000ff 
    size=2>Your theory is excellent, except for the scenario of reverse 
    splits.   There are as many reverse splits going on right now as 
    there has been normal splits in the past.  A reverse split happens when 
    a company decides that the price of their shares is too low.  Perhaps 
    they are going to be kicked out of the NASDAQ or S&P, whatever.  
    So, they reverse split the price of their shares.   I'm looking at 
    one at the moment where the actual price is only 10% of the backadjusted 
    price.    That's somewhat typical.  The price goes below 
    one dollar (say 0.80), so they reverse split the stock to make it 
    $8.00.    That makes the backadjusted price of their shares 
    $850 in 1997 instead of $85.
    <FONT face=Arial color=#0000ff 
    size=2> 
    I 
    wouldn't worry whether using actual (correct?) prices and volumes for 
    filtering is going to give you better or worse results.   It will 
    give you more accurate results.
    <FONT face=Arial color=#0000ff 
    size=2> 
    If 
    you only want to trade shares between $1 and $10 today, I assume that ten 
    years ago you would have only wanted to trade shares between $1 and 
    $10.   Why not let your systems look at the same shares in your 
    backtest that you would have considered on the day?
    <FONT face=Arial color=#0000ff 
    size=2> 
    <FONT face=Arial color=#0000ff 
    size=2> 
    <BLOCKQUOTE 
    >
      <FONT face="Times New Roman" 
      size=2>-----Original Message-----From: Phsst 
      [mailto:phsst@xxxxxxxxx]Sent: Saturday, June 21, 2003 11:15 
      PMTo: amibroker@xxxxxxxxxxxxxxxSubject: [amibroker] 
      Re: QP2 - actual and backadjusted 
      pricesChuck,Bear with me while I think 
      out loud...I have more than one trading system (long and short) 
      which producevery good backtest results (along with verified real 
      trading results)using split-adjusted data.If I revert to raw 
      (non-split adjusted) data, I'll probably see aneven better return 
      since my systems are filtered on price and volume,and I'll probably 
      pick up even more 'tech bubble' pre-split trades.And as I develope 
      even more trading systems which now use raw data, Ishould expect to 
      find it easier to ferret out profitable tradingsystems because of the 
      improved probability of 'hitting' bubblepre-split shooting stars. 
      But the reality is that by using split adjusted data, the 
      backtestresults should reveal a 'worst case' situation since 
      split-adjusteddata will eliminate many of the 'shooting stars' that 
      might beregarded as once in a decade (or even once in a century) 
      situations.It is late here... where is my thinking screwed 
      up?Phsst--- In amibroker@xxxxxxxxxxxxxxx, "Chuck 
      Rademacher"<chuck_rademacher@x> wrote:> Thanks, 
      Ed.> > Learn something new every day.   I don't 
      use QP2, but I'm sure thata lot of> people will appreciate 
      learning about the extra functionality that they> 
      offer.>   -----Original Message----->   
      From: E Winters [mailto:e.winters@xxxx]>   Sent: 
      Saturday, June 21, 2003 7:12 PM>   To: 
      amibroker@xxxxxxxxxxxxxxx>   Subject: Re: [amibroker] Re: 
      Historical volume filtering> > >   
      Chuck,>   QP2 maintains split adjusted and unadjusted 
      prices and volumes aswell.> Amibroker has a plugin for QP2 but 
      I don't think it currently has the> ability to retrieve the Raw 
      data, but there are Excel functionswhich can> retrieve the 
      unadjusted prices and volumes.>   
      Regards,>   Ed>     ----- 
      Original Message ----->     From: Chuck 
      Rademacher>     To: 
      amibroker@xxxxxxxxxxxxxxx>     Sent: Saturday, 
      June 21, 2003 1:58 PM>     Subject: RE: 
      [amibroker] Re: Historical volume filtering> > 
      >     Mornin' Al,> 
      >     My first email for the day and I haven't 
      had my cuppa yet!> >     I get my data 
      from CSI Unfair Advantage.   To my knowledge, itis 
      the> only data supplier that provides both actual and backadjusted 
      prices and> volume.    I convert the CSI data to 
      MetaStock format, placing theactual> close in the open interest 
      column.   It is possible to "squeeze"other data> into 
      the open interest column such as earnings, dividends, etc., butI'm 
      not> at the moment.   Since open interest is an integer 
      field, I multiply the> actual close by 100 before placing it in the 
      field.   If I waswriting some> AFL to filter price 
      and volume, I might say something like:> 
      >     BuyOK = OI > 100 and Volume > 
      200000;       // 100 = 
      $1>     ShortOK = OI > 600 and Volume > 
      300000;    //  600 = $6> 
      >     I'm afraid that I don't understand your 
      question about segmenting> stocks.   I simply have actual 
      and backadjusted prices all in theO,H,L,C,V> and OI fields in 
      the MetaStock data.> >     As to your 
      first question, I came up with something that works for> scaling 
      the turnover filter.   It is the most basic solution and 
      the> mathematicians would be quite critical of it.... but it 
      works:> >     LB = 
      BarIndex()-245;>     SPXVol = 
      Foreign(".SPX","Volume",fixup=1);>     
      CurrentSPXVolAvg = MA(SPXVol,245);>     
      BaseSPXVolAvg = Ref(MA(SPXVol,245),-LB);>     
      VolRatio = CurrentSPXVolAvg / 
      BaseSPXVolAvg;>     MinTurnover = 100000 * 
      (VolRatio / 3);> >     The above 
      basically calculates the 245-day moving average of 
      S&Pvolume> at any point in the data and compares it to the 
      245-day movingaverage in> the first year of data.   
      That ratio then gets divided by three(arbitrary)> and 
      multiplied by 100000.   The effect is, in 1985 I would 
      befiltering with> a minimum turnover of $100,000 and currently 
      about $500,000.> >     I hope that 
      answers your questions... now for that cuppa> 
      >         Yahoo! Groups 
      Sponsor>               
      ADVERTISEMENT> > > > >   Send 
      BUG REPORTS to bugs@xxxx>   Send SUGGESTIONS to 
      suggest@xxxx>   
      ----------------------------------------->   Post 
      AmiQuote-related messages ONLY to: 
      amiquote@xxxxxxxxxxxxxxx>   (Web page: <A 
      href="">http://groups.yahoo.com/group/amiquote/messages/)>   
      -------------------------------------------->   Check 
      group FAQ at:> <A 
      href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.html> 
      >   Your use of Yahoo! Groups is subject to the Yahoo! 
      Terms of Service.Send BUG REPORTS to 
      bugs@xxxxxxxxxxxxxSend SUGGESTIONS to 
      suggest@xxxxxxxxxxxxx-----------------------------------------Post 
      AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx (Web page: 
      <A 
      href="">http://groups.yahoo.com/group/amiquote/messages/)--------------------------------------------Check 
      group FAQ at: <A 
      href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.html 
      Your use of Yahoo! Groups is subject to the <A 
      href="">Yahoo! Terms of Service. 
      Send BUG REPORTS to bugs@xxxxxxxxxxxxxSend 
    SUGGESTIONS to 
    suggest@xxxxxxxxxxxxx-----------------------------------------Post 
    AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx (Web page: 
    <A 
    href="">http://groups.yahoo.com/group/amiquote/messages/)--------------------------------------------Check 
    group FAQ at: <A 
    href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.html 
    Your use of Yahoo! Groups is subject to the <A 
    href="">Yahoo! Terms of Service. 
    Send 
  BUG REPORTS to bugs@xxxxxxxxxxxxxSend SUGGESTIONS to 
  suggest@xxxxxxxxxxxxx-----------------------------------------Post 
  AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx (Web page: <A 
  href="">http://groups.yahoo.com/group/amiquote/messages/)--------------------------------------------Check 
  group FAQ at: <A 
  href="">http://groups.yahoo.com/group/amibroker/files/groupfaq.html 
  Your use of Yahoo! Groups is subject to the <A 
  href="">Yahoo! Terms of Service. 







Yahoo! Groups Sponsor


  ADVERTISEMENT 









Send BUG REPORTS to bugs@xxxxxxxxxxxxx
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx 
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html



Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.