Step by Step Usage:

  1. download the program, make sure you have the following folder structure in place:
   config/
   data/
   data/charts
   docs/
   src/ (you need that only if you want to compile yourself)
   tools/

Make sure you have a mysql server up and running, on your machine or on one you have access to.

  1. create a file called symbols.conf (or whatever name you like) and
    configure the symbols/markets you want to analyze:
  #symbolname = filename.csv      #comment
  DAX         =   DAX.csv         # Deutscher Aktien Index       
  DJTransp    =   DJTransp.csv    # Dow Jones Transportation Index
  DOW         =   DOW.csv         # Dow Jones Industrial Index
  EURSTOXX    =   EURSTOXX.csv    # Eurostoxx Index
  Gold        =   Gold.csv        # Gold Future
  HSeng       =   HSeng.csv       # Hang Seng Index
  1. Get price data for your markets (You can get free data from sources like finance.yahoo.com, quandl.com or investing.com)
  2. Prepare your price data, export it to csv files using the following order:
    open, high, low, close, volume, changes. Use filenames as configured in step 1.
  3. Copy the price data files to data/
  4. Copy/rename config/ichinscratchy.conf.EXAMPLE to config/ichinscratchy.conf
  5. do your configuration, make sure the # Database related settings match your mysql configuration, set CONFIG_SYMBOLFILE to symbols.conf (or however you called the file in step 1.). Take a look at docs/configuration.md for a complete description of all options.
  6. Prepare the mysql database and the user, start the program with the -s or --setup-db option:
    ./ichinscratchy -s (you can repeat this step at any given point to clean your database and start with a fresh one). Take a look at docs/database.md if you want to create your database manually (or are interested in the structure).
  7. finally run the program with or without the -v/--verbose option:
    ./ichinscratchy or ./ichinscratchy -v
  8. if configured to do so, the GNUPLOT plot files will be in data/charts, you can plot them with gnuplot plotfilename (or to plot them all, use the bash script tools/plotall.sh)
  9. With invoking the -r or --reports option you can do the statistics analysis of the generated orderbook entries, which makes sense only if you run the program over a longer time period (config option SIGNAL_DAYS_TO_EXECUTE).
  10. Play with the options, find a system that suits your needs by repeating steps 6-10
  11. Enjoy trading :)

If you are interested in seeing the program in action, visit www.spare-time-trading.de and drop me a comment or a mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

Comments powered by CComment