Value-Brief anfordern
Ergebnis 1 bis 2 von 2

Thema: Backtest-Maschinen

  1. #1
    Registriert seit
    Hattingen - Bad Kleinkirchheim


    Mechanische Strategien sind relativ einfache und oftmals risikoarme Investment-Strategien.

    Es gibt unter:

    die Jamie Gritton Backtestmaschinen. Hier können verschiedene Screens von 1986 an simuliert werden. Es ist die wohl beste Backtestmaschine.

    Es gibt eine Vielzahl von Variationsmöglichkeiten. So können z.B. Haltedauer, Starttermin, *Auswahl der gerankten Stocks, und Kaufzeitpunkt nach Ranking simuliert werden. Das Ergebnis ist eine durchschnittliche Performanceberechnung p.a. und eine Renditeberechnung für alle Monate ab 1986.

    Happy Screen !

    Die Seite enthält tolle Backtest-Maschinen:

    Standard Run
    backtests on many of the popular workshop stock screens. You can run the tests with parameters you choose, including years and starting months and number of stocks to hold. *

    Screen of Screens
    Combine the listings of multiple screens, adding their ranks and picking those that score high overall. *

    Combine different back-tested screens into a full portfolio. You can blend any screen made by one of the other backtest engines. *

    Stagger annual investments with start dates throughout the year. Can test Dozens and Dual Semi strategies, and other variations. *

    This screen selects the intersection (overlap) between two screens, with the hope that stocks that qualify for more than one similar screen (RS period, generally) will be stronger. *

    Freshmen Overlap
    Freshman stocks are those that are new on a screen, i.e. weren't on it last month (or quarter, etc). At least for RS screens, these seem to do well, especially if the overlap concept is applied to them. It's not uncommon for this screen to not find any stocks for a given month, so the regular overlap screen is used as a fallback in that case. *

    Occasional Overlap
    This applies different overlap and freshmen screens, choosing only an overlap which actually pick any stocks. *

    Rank stocks not by their position on a screen's list, but on how far up the list since they've moved since the last period. This adds a level of derivative to the rankings. *

    Buy the stocks at the top ranks of a screen, and hold them until they are no longer on the screen, or ranked too low. This includes the original "bullet" screen of buying only the #1 stock in a screen. *

    Relative Strength of Screens
    Choose from among a set of screens, picking the one that performed best in the recent past and hoping is will perform best for the present. *

    Screen Switching
    Choose between two screens based on the recent outcome of an index. Some screens may perform better in down markets, and others better in up markets; this switching would allow to invest in the best screen for the current (or at least recent) situation. *

    All-Months Summary
    Run a backtest for all possible start months at once. This removes any seasonailty effects and can give a more accurate picture of a screen's expected return. *

    Options Simulator
    This is not a true backtest, and only gives simulated results!
    Use the Black-Scholes model to simulate options purchasing strategies using the backtested screens for underlying stocks. Various combinations of long and short stock, calls, puts, and cash can be simulated. *

    Crystal Ball
    This is not a real screen!
    This meta-screen shows what return you would get if you could consitently pick next month's (or quarter's, etc) winners. This is the pure crystal-ball effect, and does not approach anything that can be replicated.

    Screen Builder
    Build your own screens from raw Value Line® data. Stocks can be selected and sorted using any combination of the supplied data fields, allowing many variations of existing screens, and well as testing completely new ideas.

    herzliche Grüße


  2. #2
    Neuer Benutzer
    Registriert seit


    Risikoarm aber nicht Risikolos. Wie sich gezeigt hat.



  • Neue Themen erstellen: Nein
  • Themen beantworten: Nein
  • Anhänge hochladen: Nein
  • Beiträge bearbeiten: Nein