Docstoc

PowerPoint Presentation - Numerical Method

Document Sample
PowerPoint Presentation - Numerical Method Powered By Docstoc
					Haksun Li
haksun.li@numericalmethod.com
www.numericalmethod.com

INTRODUCTION TO ALGO QUANT, AN
INTEGRATED TRADING RESEARCH TOOL
AN INTEGRATED SUITE OF BACK TESTING

   Data sources
   Library of signals
   Strategy templates
   Sample strategies
   Performance measures
   In-sample calibration
   Out-sample back testing
AN INTEGRATED SUITE STRATEGY ANALYSIS

   Bootstrapping
   Customized order book
   Scenario analysis
   Auto strategy generation
LIBRARY OF COMPONENTS

   Algo Quant is more than an application.
   Algo Quant is Java library of components that
    you can reuse to build your own trading
    applications, such as:
    A  customized back tester
     A quantitative strategy research tool

     An algorithmic trading system for automatic order
      execution
SUANSHU

   Algo Quant is backed by an extensive library of
    numerical algorithms for building mathematical
    trading model.
     Markov  chain
     Hidden Markov model

     Kalman filter

     Cointegration

     Regression analysis
DATA SOURCES

   Yahoo!
   Gain Capital FX rates
DATA PROCESSING

   Cleaning
   Extraction
     Equi-time

     Daily

     Weekly

   Filtering
     Moving    average
SIGNAL LIBRARY

   Open-High-Low-Close (OHLC) bar
   Arithmetic moving average
   Exponential moving average
   RSI
STRATEGY TEMPLATES

   One of the objectives of Algo Quant is that you
    can prototype a quantitative trading strategy
    very rapidly.
   Reduce the time to testing out an idea.
   Reduce the time to production.
MESSAGE BASED STSTEM

   Algo Quant is a message based system.
     event   driven
   To create a strategy, you only need to handle
    the events that concern you.
     write   handlers
SIGNAL VS. STRATEGY

   A signal takes prices (and maybe other data) to
    generate buy, sell signals, etc. It monitors and
    describes an aspect of the price process.
   A strategy, interacts with the market by sending
    orders. It determines when/what to buy and
    sell and how much.
   A strategy is a composition of signals which
    look at different aspects of the market.
PERFORMANCE MEASURES

   P&L
   Max drawdown
   Sharpe ratio
   Omega
   Your own customized measures
CALIBRATION
   Algo Quant has a suite of optimization tools to
    search for optimal parameters for a strategy
    with respect to the (historical) data for a given
    objective function.
   Optimizers:
     mixed   integer non linear programming
   Objective functions:
     Sharpe   Ratio
     Omega
BACK TESTING

   Algo Quant is a very efficient back tester as it
    runs on multiple cores.
     multipleset of parameters
     expected P&L

     variance of P&L
CUSTOMIZED ORDER BOOK

   You can customize the way an order is handled
    to simulate different execution assumptions.
     FIFO  order book
     100% execution ratio

     limit vs. market orders
COMPOSITE STRATEGY

   composite strategy = {simple strategies}
   A successful composite strategy may consist of
    not-so-successful strategies.
   A composite strategy is explainable by its
    constituent simple strategies.
   A composite strategy accounts for more market
    factors, hence more comprehensive.
SAMPLE COMPOSITE STRATEGY

   The mean reverting strategy makes small
    money most of time but loses very big money
    on trend.
   The trend following strategy loses small money
    most of the time but makes big money on
    trend.
SAMPLE COMPOSITE STRATEGY

   We combine them together to form a new
    strategy:
     run the mean reverting strategy except when there
      is an expected news/announcement event, e.g.,
      NFP.
AUTO STRATEGY GENERATION


    a strategya strategya strategya strategya strategya strategya strategya strategya strategya strategya strategya strategy




                                   search for a combination of simple strategies




                                                                                                                 add the successful strategy
                                                                                                                 to the pool so it becomes
                                                                                                                 another simple strategy




                                                     strategy verification




                                                         backtester

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:4
posted:8/14/2011
language:English
pages:19