Continuous contracts verses actual in prediction

["Rolf L. Quam" <rquam@xxxxxxxxxx>]

Those of you who have read this feed for the last couple of weeks know
that I work with genetic neural nets... I have developed some trading
systems which produce some "astonishing results" HOWEVER, all of my
analysis has been based on continuous contracts... I realize that these
are not real... I intend to add roll-over dates, current contract, next
contract, and market inversion data to my net analyisis... I would be
interested to hear from anyone who has experience with dealing with
continuous vs. actual data...

    RQ