The thing that I was really surprised about is that when I asked it to generate the swing system where average trade profit was 5% it added a few lines of AFL that used ATR to filter out relatively low volatile stocks—theory being low volatile stocks weren’t capable of generating 5% profitHoly Frack!!
I’ve only being playing with ChatGPT to generate AFL code this past week but have to say my first impressions from this week are that it’s seems reasonable at generating base code which can then be “tuned” to my requirement. It is far from perfect code, but certainly save a lot of time. I’ve found that the more specific I am with my requests the better the code is. For example, when I asked it for code with a 15% DD it inserted a stop loss and interestingly inserted code for ATR based position sizing. I also found asking it to generate the code specific to ASX (my interest) made for some interesting code variations. Was hoping to have some of the backtesting on the ChatGPT code done today but personal commitments got in the way so hoping to share result here early next week.You inspired me to finish signing up properly with ChatGPT thanks MA. Will have to play. Never would have thought to ask it to generate AFL in a million years. Impressive.
Would be interested in how you go testing its ideas - could be a new thread idea.
Separately, are your non-BO systems/MR etc doing any good in this current trading environment?
Yes, my two main live trading system are a break out and a mean reversion swing system. Unsurprisingly the Breakout system has been struggling (as most breakout systems are) but the swing system has been doing pretty well this past 3/4 months. If you’re interested I’ll post up the swing system actual live trade results for you tomorrow.
Yes please. Never had much luck with MR for ASX, so would be interested MA. Also in any figures for your ChatGPT work too. Interesting stuff
Nick Radge has been known to pop in from time to time here
Looks interesting. Like a composite index on Amibroker, or something different?Over the years I’ve created my share of indicators
They are never perfect, but they work well for me. I have observed that there is a defined correlation between the "Percentage Up Filter" above 50% and the All-Ordinaries Index rising at the same time, & restricting trading during these periods increases the chances of profitability.
2023 hasn't been kind to trend traders
In the next few posts, I'll give the signals since 1st January 2023 & you will also notice when I'm standing aside. Also, I'll post a few graphics that should be interesting & one that will be an eye-opener.
In conclusion
Incorporating metrics into the buy condition & using the observed correlation could be an essential step toward improving trading success. When the "Percentage Indicator is above 50% (the ribbon is green) it's a safe time to enter the markets & it definitely works for me.
The All Ordinaries
The chart below demonstrates the correlation between the "Percentage Up Filter" that signals a safe entry point just when the market is starting to rise.
View attachment 154724
Skate.
Great to see you back on deck Skate.View attachment 154723
It's been a while
Having other matters to attend to has resulted in the lack of time to make comments in the "Dump it Here" thread. @Nick Radge made a great Twitter post today about the logic of waiting for favorable market conditions.
Nick is reaching out for something new
There has also been a kind gesture of Nick to test the perceived logic vs the data. Well, I have such an idea, of using a "Percentage Up Filter" an indicator that I've made many posts about. I've explained the logic behind the idea & the direct correlations of this indicator in relation to the markets.
More to come.
Skate.
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?