Links work for me.
Are you on paid subscription to gpt?
Ah that is it … no, I am not on paid subscription.
Another conundrum solved, thank you.
Have anyone tried coding using the Microsoft Copilot+ PC? Is it usefull for ProBuilder?
As I understand it, its a built in Chat GPT in microsoft, or in some Microsoft tools.
Is it even possible to use the ProRealCode integration with it, or is it better to stick with the standalone Chat GPT?
ProRealCode has no integration with AI currently.
I have used AI/ChatGPT a bit lately, summarizing trading books and creating strategies from that book for me. For that use case it works really nice! Included strategies down below are from the trading book “Long term secrets to short term trading” by Larry Williams. Have fun, and please share any refinements or proper strategies based on these 🙂
// 1. 1-2-3 Low Reversal
defparam cumulateorders = false
// 1-2-3 Low setup
if low[2] < low[3] and high[1] > low[2] and low < low[2] then
buyTrigger = high[1]
endif
if not onmarket and close crosses over buyTrigger then
buy 1 contract at market
entryBar = barindex
endif
// Exit after 2 days
if onmarket and (barindex - entryBar) >= 2 then
sell at market
endif
//2. Oops! Trade
defparam cumulateorders = false
// Oops! Trade
if (open < low[1]) and (close > low[1]) then
buy 1 contract at market
entryBar = barindex
endif
// Exit: 1% profit or close
targetPrice = tradeprice(1) * 1.01
if onmarket and close >= targetPrice then
sell at market
endif
if onmarket and barindex = entryBar + 1 then
sell at market
endif
//Anti Setup (NR4 Breakout)
defparam cumulateorders = false
range = high - low
nr4 = range < range[1] and range < range[2] and range < range[3]
if nr4 then
buyTrigger = high + 0.1
sellTrigger = low - 0.1
buy 1 contract at buyTrigger stop
sellshort 1 contract at sellTrigger stop
entryBar = barindex
endif
// Exit after 2 days
if onmarket and (barindex - entryBar) >= 2 then
sell at market
exitshort at market
endif
//Monday Reversal
defparam cumulateorders = false
mondayDown = (dayofweek = 1 and close < open)
if mondayDown and dayofweek = 2 and close > open then
buy 1 contract at market
entryBar = barindex
endif
// Exit after 2 days
if onmarket and (barindex - entryBar) >= 3 then
sell at market
endif
//Inside Day + Narrow Range
defparam cumulateorders = false
Myrange = high - low
insideDay = high < high[1] and low > low[1]
narrow = Myrange < Myrange[1]
if insideDay and narrow then
buyTrigger = high + 0.1
sellTrigger = low - 0.1
buy 1 contract at buyTrigger stop
sellshort 1 contract at sellTrigger stop
entryBar = barindex
endif
// Exit after 2 days
if onmarket and (barindex - entryBar) >= 2 then
sell at market
exitshort at market
endif
I have used AI/ChatGPT a bit lately, summarizing trading books and creating strategies from that book for me. For that use case it works really nice! Included strategies down below are from the trading book “Long term secrets to short term trading” by Larry Williams. Have fun, and please share any refinements or proper strategies based on these 🙂
// 1. 1-2-3 Low Reversal
defparam cumulateorders = false
// 1-2-3 Low setup
if low[2] < low[3] and high[1] > low[2] and low < low[2] then
buyTrigger = high[1]
endif
if not onmarket and close crosses over buyTrigger then
buy 1 contract at market
entryBar = barindex
endif
// Exit after 2 days
if onmarket and (barindex – entryBar) >= 2 then
sell at market
endif
// I did a backtest and liked this code. I have added a short and did some updates. it shows promis and I will backtest on live data for a few days and see if I can refine it a bit more. 1H Timeframe
// === SETTINGS ===
DEFPARAM CumulateOrders = False
// === VARIABLES ===
if barindex = 0 then
entryBar = 0
buyTrigger = 0.0
sellTrigger = 0.0
endif
// === 1-2-3 LOW REVERSAL SETUP (LONG) ===
if low[2] < low[3] AND high[1] > low[2] AND low < low[2] THEN
buyTrigger = high[1]
endif
// === 1-2-3 HIGH REVERSAL SETUP (SHORT) ===
if high[2] > high[3] AND low[1] < high[2] AND high > high[2] THEN
sellTrigger = low[1]
endif
// === LONG ENTRY ===
if NOT onmarket AND close crosses over buyTrigger THEN
buy 10 CONTRACT AT MARKET
entryBar = barindex
endif
// === SHORT ENTRY ===
if NOT onmarket AND close crosses under sellTrigger THEN
sellshort 10 CONTRACT AT MARKET
entryBar = barindex
endif
// === LONG EXIT ===
if LONGONMARKET AND (barindex – entryBar) >= 20 AND close > tradeprice THEN
sell AT MARKET
endif
// === SHORT EXIT ===
if SHORTONMARKET AND (barindex – entryBar) >= 20 AND close < tradeprice THEN
exitshort AT MARKET
endif
24 Hours of Live Testing the above Code. it has promis and will continue to Live Test in my Demo Account.
Fun to see the strategy looks promising! I cant seem to reproduce your results. Do you mind uploading .ITF per instrument?
On my ChatGPT Account I have a link in the sidebar which when I click it shows this …
ProRealTime Code
A software engineer expert in ProRealTime platform coding and formatting. GENERAL USE.
Apart from me seeing above and after I click the link … chat GPT appears to work for me exact same as if I DO NOT click on the autotrading.ee link in the sidebar.
My question is:
What is the idea / function of this link / autotrading.ee sponsored chat GPT page (?) and how is it different to me just opening my own chat with GPT (i.e. without me clicking on the autotrading.ee linked ChatGPT page).
Hi Grahal, that’s from your friend here (see below).
Ask him how he got that in there.
Anyway it will just be spam.
You didn’t sign up with GPT so it can fire up its own computing, did you ? (this is new since a few days)
JSParticipant
Senior
GPTs are customized versions of ChatGPT that have additional information, knowledge, instructions, and skills about a specific subject — for example, ProRealTime… 😉
In ChatGPT, you can create these custom versions yourself…
You can also search under “Explore GPTs” for specific topics (see screenshot)…
When you use these specialized GPTs, the answers will generally be more relevant and contain fewer errors, as the GPT is already “trained” on the specific topic…
Currently, there are 8 GPTs (in the free version) related to ProRealTime…
Just to clear up a common mix-up 🙂: custom GPTs in ChatGPT aren’t re-trained models.
They’re the same underlying ChatGPT, but with added instructions, reference docs, or even API hooks.
It works a bit like RAG (retrieval-augmented generation) when you ask something, the GPT first looks through that extra material you’ve provided, then uses its usual reasoning skills to create the answer.
That’s why they can feel more accurate on a topic and not because the AI “learned” it from scratch, but because it’s got the right notes in front of it.
I’m currently working on two projects around ProRealTime coding:
- ProRealBlocks – lets you build ProBuilder code by assembling visual blocks. The code is generated in real time as you connect them. Perfect for learning how to code through block-based reasoning. This one’s always ready (for the new website launch) and currently delivers basic code reasoning based on its built-in knowledge (see attached examples).
- An advanced ProRealTime coding chatbot – designed to be the most accurate bot for generating ProRealTime code. Still deep in development (lots of trial and error! and model training can take a lot of time..), but very promising. There’s still a lot to do before it’s production-ready.
Fun to see the strategy looks promising! I cant seem to reproduce your results. Do you mind uploading .ITF per instrument?
Here is the .ITF and a printed P/L from the 5th of Aug showing untouched transactions. Also attached Current position, I’m running it on a 5 minute TF and will let it run for 30 days
