Does really nobody know this?

P

PietdeWit

I want to backtest intraday systems on ratio-based back-adjusted future contracts with TradeStation. Who knows what firm offers this kind of back-adjusted intradaydata? CSIData and TickData cann't offer me this. Point-based back adjusted data is easy to find buth that is not where I'm looking for. Who gives me a clue?
 
Why do people like to backtest historical data? I think this is an inherently flawed process that has no bearing on the future.
 
Originally posted by PietdeWit
I want to backtest intraday systems on ratio-based back-adjusted future contracts with TradeStation. Who knows what firm offers this kind of back-adjusted intradaydata? CSIData and TickData cann't offer me this. Point-based back adjusted data is easy to find buth that is not where I'm looking for. Who gives me a clue?
More...

I do not see any sense in testing intraday systems with adjusted contracts. Just splice the contracts. i.e. append to the end of another and take care of the rollover days when you use any indicators.

The raison d'etre for adjusted contracts is in EOD -Systems which might deliver wrong results when using non-adjusted data.

But intraday? Why?

Have fun

Bernd Kuerbs
 
Originally posted by goldenarm
Why do people like to backtest historical data? I think this is an inherently flawed process that has no bearing on the future.
More...

I'd prefer to backtest data from the future, but it's even harder to find. :D :D :D
 
Originally posted by ddefina


I'd prefer to backtest data from the future, but it's even harder to find. :D :D :D
More...

Hmm, are we allowed to call that backtest? Shouldn't it be called forward test?

I agree, it's the only sensible way of testing, but the data vendours refuse to sell that data, though they could make lots of money...

Maybe there is some hope in utilizing the recently tested effects of quantum coupling to get access to that data?

Ok, ok, it was a long hot day, I'm going to bed now....

regards

Bernd Kuerbs
 
Originally posted by goldenarm
Why do people like to backtest historical data? I think this is an inherently flawed process that has no bearing on the future.
More...

because it can help describe how certain actions worked under certain conditions. As long as it's seen as a tool in that sense, it's very useful. If it's used to sell systems, it's questionable.
 
You must log in or register to reply here.
Back
Top