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

RE: [amibroker] Re: TRENDING vs. RANGING markets



PureBytes Links

Trading Reference Links




<FONT 
color=#000080>Hilbert?
How about <SPAN 
class=451545321-27032003>MaximumEntrophySpectralEstimation? 

<SPAN 
class=451545321-27032003>Bettter yet, call it <SPAN 
class=451545321-27032003>MaximumEntrophySpectralAnalysis to 
attract the MESA crowd.
<SPAN 
class=451545321-27032003> 
<SPAN 
class=451545321-27032003>Bob

  <FONT face=Tahoma 
  size=2>-----Original Message-----From: Listes trading 
  [mailto:listes.trading@xxxxxxx]Sent: Thursday, March 27, 2003 5:16 
  AMTo: amibroker@xxxxxxxxxxxxxxxSubject: RE: [amibroker] 
  Re: TRENDING vs. RANGING markets
  <FONT face=Verdana color=#800000 
  size=2>Dimitri
  <FONT face=Verdana color=#800000 
  size=2> 
  If 
  I have clearly understood the MESA documentation, MESA does not use FFT but 
  another algorithm that is not subject to the FFT 
  limitations.
  <FONT face=Verdana color=#800000 
  size=2>FFT has 2 major limitations for financial 
  applications:
  
    <FONT face=Verdana color=#800000 
    size=2>FFT is valid only for time series having a finite second order moment 
    and there is no evidence that financial time series fulfil this condition 
    (some theorician like B.Mandelbrot has made the hypothesis of infinite 
    moment for market time series) 
    <FONT face=Verdana color=#800000 
    size=2>FFT uses moving windows and is very sensitive to the choice of the 
    window and to the way data are padded, this is why - as you state- the 
    historical results are changed when you add new data
  I 
  think MESA is using Hilbert transform that is very efficient to analyse 
  signals having a single frequency varying in time. It is able to extract the 
  variation of the frequency. But when the signal contains several frequencies 
  varying in time, the results are not reliable. Financial time series are 
  composed of an unknown number of frequencies varying in time and, therefore, 
  Hilbert transform does not seem very suitable for analysis and 
  prediction.
  <FONT face=Verdana color=#800000 
  size=2>Anyway, MESA site contains a lot of interesting documents and even if 
  it is not Holy Grail, it is a great source of ideas to built indicators and 
  systems.
  <FONT face=Verdana color=#800000 
  size=2> 
  <FONT face=Verdana color=#800000 
  size=2>Waz
  
    <FONT face=Tahoma 
    size=2>-----Original Message-----From: DIMITRIS TSOKAKIS 
    [mailto:TSOKAKIS@xxxxxxxxx]Sent: jeudi 27 mars 2003 
    11:30To: amibroker@xxxxxxxxxxxxxxxSubject: [amibroker] 
    Re: TRENDING vs. RANGING marketsGreg,thank you 
    for your mail.I always have a question for any spectral study. They are 
    usually based of FFT, which changes the historical results as time goes 
    by.My specific question is for the Cycle function of your gif. Suppose 
    we read a price 20 for March27. Three months later, the reading of 
    March27 will still be 20 or something else ?TIA for any 
    reply.Dimitris Tsokakis--- In amibroker@xxxxxxxxxxxxxxx, "Greg" 
    <greg.bean@xxxx> wrote:> Gosub,> > I have just 
    recently been doing some reading about this in John Ehler's -> 
    Rocket Science for Traders . He does make an attempt at "predicting" 
    future> price movement. Below is a brief example of his approach to 
    this. I haven't> found out how to test this in AB , but maybe 
    someone else has.> > <A 
    href="">http://www.mesasoftware.com/mesa98.htm#MESA96> 
    > MESA2002:> > MESA2002 is THE premier cycle-based 
    trading program. MESA2002 is available> for a variety of trading 
    platforms, includingTradeStation (4.0, 2000i, and> 6.0), 
    SuperCharts, NeuroShellTrader, and Standalone. It accurately 
    measures> short term cycles. The trading platforms offer powerful 
    extensions to> trading systems by making variables adaptive to 
    the measured cycles. One> such adaptive application is to 
    determine whether the mode of the market is> cyclical or 
    trending.> > > MESA2002 program Charting Screen> 
    > The basic operation of MESA2002 is measurement of the spectral 
    content of> the price data using the Maximum Entropy Spectral 
    Analysis method. When the> data is stationary the cycles are 
    stable and consistent. The spectral> display segment is a 
    colorized contour plot showing the quality of the cycle> 
    measurement (the bottom display segment). When the measured cycles 
    are> erratic or when the spectral energy is "splattered" across 
    the range of> cycle periods the market is in a Trend 
    Mode.> > > > > The market mode is sensed 
    by examining the phase of the dominant cycle. A> fundamental 
    definition of a cycle is a constant rate change of phase. For> 
    example, a 10 day cycle changes phase at the rate of 36 degrees per day 
    to> complete the 360 degrees in each cycle. The Trend Mode is 
    identified by the> failure to change at a constant rate. The 
    phase presentation in the 3rd> display segment shows the phase 
    changing at a constant rate, forming a> sawtooth waveform in the 
    Cycle Mode, and changing erratically in the two> Trend Mode 
    areas.> > The Sinewave Indicator in the 2nd display segment 
    gives the position> reversal signals by the crossing of the two 
    lines. Note these two lines> cross only when the prices are in 
    the Cycle Mode and, unlike most oscillator> signals, do not give 
    false whipsaw signals when the price is in the Trend> Mode. The 
    Sinewave Indicator is anticipating the cyclic turning points, is> 
    generated by adding 45 degrees to the measured phase angle and plotting 
    the> Sine of it.> > The strength and direction of the 
    Trend are shown by the two adaptive moving> averages overlaid on 
    the price bars. The slower of these averages is an> Instantaneous 
    Trendline, obtained by completely removing the dominant cycle> 
    component. The faster of these averages is a minimum lag filter.> 
    > MESA2002 (standalone and TradeStation versions) makes a prediction 
    of prices> 10 bars into the future. This prediction is made on 
    the assumption that the> measured dominant cycle will continue 
    into the future with the same> amplitude and phase. Therefore, the 
    prediction has greater validity when the> prices are in the Cycle 
    Mode.> > MESA2002 (standalone) reads five different data types 
    directly, including> Metastock, CSI, ASCII, and TC2000. You can 
    generate your own custom> portfolio and have MESA2002 automatically 
    scan that porfolio for buying and> selling opportunities.> 
    > > > The MESA2002 DEMO is a 534 KB self extracting 
    file. When you run MESADEMO,> it is extracted to the C:\MESADEMO 
    directory as the default. To run the> demo, you click START . . . 
    RUN and then type C:\MESADEMO\DEMO32. Then click> on the 
    MESA2002.DBD file.> > ...........................MESA2002 
    (standalone) .........$350> ...........................MESA2002 for 
    NeuroShell Trader. .........$350> 
    ...........................MESA2002 for TradeStation2000i or 6.0> 
    .........$495> ...........................MESA2002 for TradeStation 
    4.0. .........$495> ...........................MESA2002 for 
    SuperCharts. .........$250> > . . . . . .> > 
    > > (Back to Start)> > > > 
    > ----- Original Message -----> From: "gosub283" 
    <gosub283@xxxx>Gosub> To: 
    <amibroker@xxxxxxxxxxxxxxx>> Sent: Wednesday, March 26, 2003 
    1:08 PM> Subject: [amibroker] TRENDING vs. RANGING markets> 
    > > > Hi everyone,> >> > I think this 
    issue will become more important> > over the next year or 
    two.....> >> > When a human looks at a chart, he/she 
    can> > immediately determine if a market is in> > a 
    TRENDING mode or a RANGING mode.> > It is a most amazing feat of 
    human visual> > data analysis that takes place in a matter> 
    > of seconds. Trying to get computers to "visually"> > analize 
    anything takes major computing power.> > Unfotunately computers 
    and trading system have> > a much more difficult time of 
    determining these> > market modes than us humans.> > 
    Things become very "fuzzy" when trying to> > put the question of 
    "Ranging vs. Trending" into> > mathamatical algorithms. And...of 
    course.. the> > trading timeframe (long term vs. short 
    term),> > make things even fuzzier.  (is "fuzzier" a word 
    ??)> >> > For those of us who program automated 
    systems,> > this is especialy important because it means> 
    > that we can design systems that adjust effortlessly> > 
    between ranging and trending markets. If correctly> > identified, 
    a system can use a particular set> > of indicators for Ranging 
    markets and then switch> > to other indicators when a trend is 
    determined.> > Allowing for a truly autonomous system.> 
    >> > Has anyone found a way to mathamatically 
    (reliably!)> > determine if a market is Trending or Ranging 
    ???> > (An AFL algorithm perhaps)> > In other words, an 
    indicator which can give direction> > as to which set of 
    indicators to use.> > (A hypothetical example of such an indicator 
    would> >  have a response from 0=ranging to 10=Trending 
    )> >> > Cheers,> > Gosub283> 
    >> >> >> > 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 <A 
    href="">http://docs.yahoo.com/info/terms/> 
    >> >> >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.