Great videos, Great channel. The EAs you've shown on the channel have extremely low trade frequencies. Your sample sizes do not have any statistical significance. I would like to ask, whats the minimum number of trades, per TF, per currency pair for a buy or sell setup do you look for in a backtest? Because 30trades, 100trades or even 1000trades are insignificant. U could be trading a random pattern. I really love your channel, I hope you won't take this as an attack.
Hi There, not at all, I take no offense and welcome comments. CRYPTO is too young to form any sort of statistical significance. If you implement a trend following system from the day bitcoin started trading on the daily timeframe you will never get statistical significance. Check this article out these backtests where run from the day bitcoin started and only 133 trades where made. www.quantifiedstrategies.com/trend-following-and-momentum-on-bitcoin/ www.quantifiedstrategies.com/bitcoin-and-crypto-guide/ So you really just finding a good strategy and will continue with forward optimizations. if you get a trade everyday since 2009 thats only 4 745 trades.
@@AlgoTrading Thanks for the response. How many trades in currency pairs do you prefer to see, not in crypto, that will made a back test significant, according to your personal experience testing and trading algo strategies. Because with currency pairs you can get good data with no gaps as far as 1998.
@@yamkelanjikelana6261 If I get good results with 10000 trades I am normally happy with that. I then like to see at least 5000 trades on OUT OF SMAPLE data.
![Blox Fruits Dragon Rework Update [Full Stream]](http://i.ytimg.com/vi/EqDAp8udhm0/mqdefault.jpg)
![Noob To Max With DRAGON REWORK In Blox Fruits [FULL MOVIE]](http://i.ytimg.com/vi/LnBMOinoOvA/mqdefault.jpg)
Great videos, Great channel. The EAs you've shown on the channel have extremely low trade frequencies. Your sample sizes do not have any statistical significance. I would like to ask, whats the minimum number of trades, per TF, per currency pair for a buy or sell setup do you look for in a backtest? Because 30trades, 100trades or even 1000trades are insignificant. U could be trading a random pattern. I really love your channel, I hope you won't take this as an attack.
Hi There, not at all, I take no offense and welcome comments. CRYPTO is too young to form any sort of statistical significance. If you implement a trend following system from the day bitcoin started trading on the daily timeframe you will never get statistical significance. Check this article out these backtests where run from the day bitcoin started and only 133 trades where made.
www.quantifiedstrategies.com/trend-following-and-momentum-on-bitcoin/
www.quantifiedstrategies.com/bitcoin-and-crypto-guide/
So you really just finding a good strategy and will continue with forward optimizations. if you get a trade everyday since 2009 thats only 4 745 trades.
@@AlgoTrading Thanks for the response. How many trades in currency pairs do you prefer to see, not in crypto, that will made a back test significant, according to your personal experience testing and trading algo strategies. Because with currency pairs you can get good data with no gaps as far as 1998.
@@yamkelanjikelana6261 If I get good results with 10000 trades I am normally happy with that. I then like to see at least 5000 trades on OUT OF SMAPLE data.
Seriously struggling with EA, thanks for your efforts. The terms used are really strange dou I know it cos am new to EA