Strategy evaluation steps?

K

Konradp

Hi,
I’d like to ask what steps You take while testing Your trading strategy.
Of course backtesting, monte carlo, other common steps but maybe there is something extra You use. I recently found that testing strategy with different starting date helps a lot in determining how it behaves.
 
Konradp said:
Hi,
I’d like to ask what steps You take while testing Your trading strategy.
Of course backtesting, monte carlo, other common steps but maybe there is something extra You use. I recently found that testing strategy with different starting date helps a lot in determining how it behaves.
More...
I spend my time trying to prove it doesn't work.
 
1) Optimization and backtesting
2) Backtesting with out-of-the-sample dataset (data not included in previous step)
3) If results are good, analyze if forward data was use in 1 and 2 (or any other type of error)
3) Paper trading (real-time)
4) Real-money trading with small volume
5) Real-money trading with full account $$

Most of people go from 1 to 5, and fail.
 
Start of with a 3 step process
  • First watch your educational content once
  • Second time watch and take notes and do some testing on the charts (live and bactesting)
  • Third time watch it and pick up all the details you missed while reading the charts

Then start by building a trade plan
its okay to trade a small capital at the start to understand emotions
80% emotions while trading

Then when you start back-testing build a data sample
eg: number of trades, time of the day, note down what time you faced the most losses, note down your emotions while trading live and understand why it happened, have a spread sheet to keep all the data, SR and RR.

Once you have enough data to go live then you start
live testing on a small live account and see how it work
collect data( the market is never the same, so don't just rely on the back-tested data)

Then its a constant Refine, Refine, Refine process, till the end.
 
rb7 said:
1) Optimization and backtesting
2) Backtesting with out-of-the-sample dataset (data not included in previous step)
3) If results are good, analyze if forward data was use in 1 and 2 (or any other type of error)
3) Paper trading (real-time)
4) Real-money trading with small volume
5) Real-money trading with full account $$

Most of people go from 1 to 5, and fail.
More...

And by the time you get to #5, your system will have run its course as evidenced by an immediate drawdown...and you get to start all over again.
 
You must log in or register to reply here.
Back
Top