The opinions expressed in these forums do not represent those of C2, and any discussion of profit/loss
is not indicative of future performance or success.
There is a substantial risk of loss in trading. You should therefore carefully consider
whether such trading is suitable for you in light of your financial condition. You should read,
understand, and consider the Risk Disclosure Statement that is provided by your broker
before you consider trading. Most people who trade lose money.
I would like to take this opportunity to provide an update in light of the current circumstances. We find ourselves in a significant drawdown period, that has resulted in questions/comments/etc, which warrants an update.
Basically, system will do what it does and it is working as designed. Note below points:
System Stop Mechanism: Our system is equipped with a daily stop feature, which serves to mitigate risk on a daily basis. However, we have refrained from implementing a “system stop” based on our rigorous testing. Every attempt to introduce such a stop has resulted in adverse effects on the system’s overall performance.
Independent Daily Decisions: It is crucial to understand that each trading day should be treated as an independent event, irrespective of past performance. When considering the system’s decisions, it is prudent to adopt the perspective of a new user just joining our platform. What would be the end-of-day (EOD) decision for this new user? This is the lens through which we evaluate our system’s actions. Therefore, even if the system has experienced losses over consecutive days, the EOD decision for today, as of September 21, 2023, is to be in a long position (going into Friday 22nd).
You maybe wondering, “if system has lost 2, 3, 4 days in a row, could that mean system can’t handle current situation”? If so a system stop, or “going opposite”, would help if this is true. I’ve tried system stops (point #1), and “going opposite” after a X (tried various X) days in a row, none of which helps. If anyone has ideas let me know but I’ve tried a lot of mine and user’s ideas and suggestions.
Hi Tony,
Thanks for the update. I am a bit confused here: even as a new user join C2 and make decision to the trade, he/she will still take previous days market into consideration; but it sounds like for your system, it will only evaluate market of current day. Is my understanding correct?
Welcome! For your question, the system takes the past X days into account, but there’s some info up to months ago that’s still taken into account. What I mean is, each day’s decision is independent. You can run the system today, without having ran it yesterday.
Say there’s two users:
UserA: Was short today and won big
UserB: Was long today and lost big
For both of them, the system will run independently of the fact that one was short and one was long. It ignores that. It decides TQQQ (long) is the choice at EOD, and both userA and userB will make their account to be long (TQQQ).
The model doesn’t take as input, if the user is up a lot, or down a lot last few days. Basically the model doesn’t consider as input the fact that it’s past X predictions is right or wrong. One can argue that it should, and I’ve tested that, but it always resulted in worse performance.
Thank you for your response. As a new subscriber to your system, I have two further questions for you:
How frequently do you make adjustments or fixes to the system? As you mentioned,
If you observe that the model consistently predicts correctly or incorrectly over an extended period, would you intervene and make adjustments to the system?
2. Leveraged ETFs experience significant decay during volatile market conditions. What measures has your system taken to mitigate this effect? Especially since I’ve noticed that some of the overnight positions are held for relatively long periods.
Yes, checks are done to make sure everything running ok (this is done anyways periodically). Also, system will be manually checked for possible adjustments to run better (this isn’t always possible, but sometimes is).
Nothing is built in, manually, specifically to handle the decay. The idea is for the AI model to take the decay into account when training, on it’s own. This results in holds for relatively long periods at times.
