Re: XML Feeds?
Sommertime - I tried to develop own arbitrage program and I could send you some information about the implementation problems. In my experiment I used program language Java, database HSQLDB and OS Ubuntu (free technologies). This is very heavy task because you should create a database with teams, games, roi of each issue. For example if some team is 'FC Barcelona' in site A, and 'Barcelona' in site B, your program should be able to detect that this is only 1 team. The same problem remains with the names of the games. So you should develop specific program interface (xml parser) for each bookmaker and store the different information from the xml feeds in one database with 1 standard format. Then you will be able to run an arbitrage algorithm and find the events with maximal roi. I think this is the main advantage of the xml feeds - you could use general algorithms from the Linear algebra, Mathematical logic and Mathematical optimisation (Simplex method) to find the events with maximal roi. This is not possible if you search manually some events directly from the site of the bookmaker.
|