you can send me an email: bptrades@gmail.com and I can see what
you are looking out, can't really comment until I know what you are
looking at.
strategies can be done in a many ways. For the SPY and ES system
strategies since they are daily strategies my approach there was to
test as far back as possible finding settings that worked in all
market conditions extreme bull, bear, sideways etc - I wanted to
get as much data as possible. For day trade strategies that's not
generally very feasible because of the sheer volume of data - like
if you try and backtest a system setting on a 5 min chart going
back to 1995 - your test might take days or weeks. For intra day
strategies the best thing is to either backtest every couple weeks
or every weekend so that you are always adjusting to the current
market. Another method that is a lot more complicated but it makes
a lot of comment sense to me would be to look at past movements and
run multiple backtests for different market environments. And
example would testing conditions when the market was in a strong
uptrend staying above the 20 day MA. Then test another set of
conditions where price is sideways. Then test another set of
conditions where the market is downtrending. You would get
multiple variations of settings for your strategy for different
market conditions and run the one that fits the current market
conditions. There's some subjectively doing this but it's a logical
method vs trying to curve fit one set of conditions over many
years.
you can send me an
Matt, I'm making my own tradestation system and it was ...
Posted by matt on 26th of Jun 2022 at 03:40 pm
you can send me an email: bptrades@gmail.com and I can see what you are looking out, can't really comment until I know what you are looking at.
strategies can be done in a many ways. For the SPY and ES system strategies since they are daily strategies my approach there was to test as far back as possible finding settings that worked in all market conditions extreme bull, bear, sideways etc - I wanted to get as much data as possible. For day trade strategies that's not generally very feasible because of the sheer volume of data - like if you try and backtest a system setting on a 5 min chart going back to 1995 - your test might take days or weeks. For intra day strategies the best thing is to either backtest every couple weeks or every weekend so that you are always adjusting to the current market. Another method that is a lot more complicated but it makes a lot of comment sense to me would be to look at past movements and run multiple backtests for different market environments. And example would testing conditions when the market was in a strong uptrend staying above the 20 day MA. Then test another set of conditions where price is sideways. Then test another set of conditions where the market is downtrending. You would get multiple variations of settings for your strategy for different market conditions and run the one that fits the current market conditions. There's some subjectively doing this but it's a logical method vs trying to curve fit one set of conditions over many years.
Thanks for the feedback...I'll pop
Posted by mrjasongill on 27th of Jun 2022 at 11:03 am
Thanks for the feedback...I'll pop you an email.
Matt, when you backtest, do
Posted by bulf6285 on 27th of Jun 2022 at 06:40 am
Matt, when you backtest, do you test for Net Profit, Winning%, other?