データ構築戦略
3件の返信
Hi everyone
what is the difference between build strategy from 2013-2018 (50%is 50% oos) and for example 2003-2018 (25-30% oos) ?
I risk of generating a overfitting strategy?
ありがとう
0
こんにちは,
some say the there is no point using data from 20 years ago. I believe the more data the better. If I can find a strategy that made money steadily for 20 years it tells me its worth to explore. You can easily find a strategy that made money for 7 years but failed for other 13 years.
Keep in mind you should always have some separated portion of data (like last 2 or 3 years) that is used to evaluate historical performance trend. If the strategy continues to perform in a similar fashion it is again a very good clue it could continue performing in following years …
0
Okay, you’re right.
Do you think it makes sense to use WFM to improve profits in the near future?
Maybe for only strategy that only earn in recent years
0
every user could use different workflow – some of thems using WFM only as a robustnest check, so they are not using optimised parameters for the future, some of them does…etc. etc.
for me its a question of your workflow and your experience – how do you set the WFM, how do you evaluate it and what you will do with the info provided from WFM
there is nothing like “best way”
利益を上げるアルゴトレーダーになりたいですか?私たちは2014年序盤からStrateQuantソフトウェアを使い始めました。現在では、あらゆるタイプの市場に対応するEA構築に関する膨大なノウハウを持っています。私たちはこのノウハウ、アプリ、ツール、そしてすべての最終的な戦略をリアル・トレーダーの皆様と共有しています。参加したい方は、以下にご記入ください フォーム.
0
3件の返信を表示中 - 1 - 3件目 (全3件中)