billpwchan / futu_algo Goto Github PK

Hi,

Would be nice to have a feature that sets up a cron job in Github Actions to process all stocks using a pre-defined strategy and send out the daily selected stocks to subscribers. The subscribers can simply include their name in a list or comment them in GitHub and you can setup a workflow to automatically update that list on a daily basis.

Let me know what's your thought! 😊

Add support for using Seaborn / Matplotlib to plot the trading result.

Will it support US Market

Is your feature request related to a problem? Please describe.
No

Describe the solution you'd like
After receiving the data, use another thread with timeout to place buy/sell orders.

Describe alternatives you've considered
N/A

Additional context
N/A

https://docs.python.org/2/library/functions.html#getattr

Is your feature request related to a problem? Please describe.
N/A

Describe the solution you'd like
It would be better if futu_algo can support Options Trading Strategies such as bull call spread in addition to the existing spot trading

Describe alternatives you've considered
Interactive Brokers might have a more mature solution for US market, but for HK stock market, it would be easier to just develop a simple options trading engine from scratch.

Describe the bug
FileNotFoundError: [Errno 2] No such file or directory: 'D:\Users\admin\Documents\Python\futu_algo-master\data\Stock_Pool\ListOfSecurities.csv'

Describe the bug
When execute the main_backend.py file，it return:

usage: setup.py [global_opts] cmd1 [cmd1_opts] [cmd2 [cmd2_opts] ...]
or: setup.py --help [cmd1 cmd2 ...]
or: setup.py --help-commands
or: setup.py cmd --help

error: no commands supplied

the setup.py --help does not work

what is the cmds, may i have the example of the parameters

if possible, we can use Chinese to communicate

Hi there. I'm trying to run the main_backend.py but I keep getting an error:

ERROR — Cannot subscribe to K-Line: No right to subscribe 00001

I have a subscription quota of 300 according to OpenD and subscription status query also shows 300:

INFO — Query Subscription Quota: {'total_used': 0, 'remain': 300, 'own_used': 0, 'sub_list': {}}

Does anyone know why this might be the case?

Any advice appreciated. Thanks.

• Futu version 7.1.3308

Is your feature request related to a problem? Please describe.
No

Describe the solution you'd like
Currently, the script only works if the user buys Lv2 data subscription card with 318HKD / Month. However, if the IP address is from Mainland China, FuTu automatically grant Lv1 data subscription without extra cost. However, we don't have Orderbook and some other features comparing with Lv2.

Need to update the solution and another configuration field in Config.ini to indicate which data subscription plan is used.

Describe alternatives you've considered
No

Additional context
No

Use 1M interval stock data to quickly calculate other time interval-based stock data on demand.

Potential useful resources https://stackoverflow.com/questions/62373692/how-do-you-convert-1-minute-open-high-low-close-data-to-another-timeframefx-5

Is your feature request related to a problem? Please describe.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]

Describe the solution you'd like
A clear and concise description of what you want to happen.

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.

Additional context
Add any other context or screenshots about the feature request here.

When converting 1M stock data to any time interval greater than 60M, it creates discrepancy between the generated file & actual file due to the non-trading hours.

For instance, for 1H interval, the suggested key timestamps are

1. 9:30
2. 10:30
3. 11:30
4. 12:00 -> Creating issue because the difference is only 30M between this one as the previous timestamp

Need to update the logic to support this types of conversion using 1M-based stock data.

Instead of subscribing to the data and rely on the data push from the Futu side, use multi-threads to retrieve the latest data from the Futu OpenD with no frequency limit.

Check out https://openapi.futunn.com/futu-api-doc/qa/quote.html for more details.

The implementation should take around 2 weeks to finish. It's a major one, should be tested thoroughly before merging to the Master

1. Remove all dependencies in trading_engine.py, and move all data processing scripts into data_engine.py
2. Use Pathlib instead of file path string for cross-platform support.

1. Refactor the code so all the I/O related operations are encapsulated in data_engine
2. Convert CSV to Parquet with 2X saving in file size...Amazing
3. Create a utility function for ppl who've already downloaded historical data in CSV format, and they can easily use the utility function to convert all CSV to Parquet ease.

Instead of creating an instance for each stock code, re-use strategies instance for stocks that are configured to use the same strategy.

Do you have plans to support US options?
Thank you

Is your feature request related to a problem? Please describe.
Nah

Describe the solution you'd like
Update HFT Stock Pool based on a separate stock selector (using primarily Yahoo Finance data).

Describe alternatives you've considered
Support other data sources such as Google Finance or other paid subscribed data source.

Additional context

ADX & RSI Strategy based on the following specification

ADXdays=Optimize("ADXdays",10,5,25,5);
ADXlimit=Optimize("ADXlimit",25,10,50,5);
RSI_threshold = Optimize("RSILimit",20,5,100,5);

Buy= RSI(2) < RSI_threshold AND ADX(ADXdays)>ADXlimit ;
buyPrice=Close;

Sell= C>Ref(H,-1);

In Testing, each buy/sell decision is logged twice for an unknown reason.

deployment futu_algo step by step , but when execute , exception occured. how to fix? thanks
(base) /home/xiaoming/anaconda3/bin/python3 main.py
Please rename stock_strategy_map_template.yml to stock_strategy_map.yml and update it.
Traceback (most recent call last):
File "/home/xiaoming/develop/futu_algo/main.py", line 31, in
from engines import *
File "/home/xiaoming/develop/futu_algo/engines/init.py", line 22, in
from .order_engine import *
File "/home/xiaoming/develop/futu_algo/engines/order_engine.py", line 21, in
from futu import ModifyOrderOp, OpenHKTradeContext, OpenQuoteContext, OrderStatus, OrderType, RET_OK,
ImportError: cannot import name 'ModifyOrderOp' from 'futu' (/home/xiaoming/anaconda3/lib/python3.9/site-packages/futu/init.py)

Is your feature request related to a problem? Please describe.
No

Describe the solution you'd like
When we have a buy/sell signal, send an Email / SMS/ WhatsApp Message to clients

Describe alternatives you've considered
Nah

Additional context
Nah

Hello

Great project you've got here! I believe Futu provides support for paper trading, Does this support paper trading though?

Is your feature request related to a problem? Please describe.
No

Describe the solution you'd like
So currently, the strategy might lead to incorrect buy decisions for some stocks with tiny transaction volumes (e.g., a sudden large buy from funds will cause the technical indicators to suddenly satisfy the condition of the buy strategy). It is better to filter out these less healthy stocks in our stocks pool before starting the high-frequency trading

Describe alternatives you've considered
May be implemented before the trading starts or after the trading starts.

Additional context
Nah

Feel free to leave your thoughts!! Any name suggestion is welcomed

It seems some stock's data aren't correct when I downloaded them. For instance, for stock HK.01997, the 1M data I downloaded isn't consistent with the one I observed in the Futu client. Gotta use data validator to ensure all data are correct before backtesting.

Hi there, is it possible for the project to run on Mac OS or Linux system? Or any plan for this?

StockQuote Handler Base will push a message to subscriber everytime the price is changed.
RTData Handler Base will push a message every 1 minute.

To use strategies like MACD crossover, it is better to use RTDat than StockQuote which is less sensitive in real-time.

Feature request
User login to our website and pay. We add them to the list for sending email and SMS.

For each stock, use a different strategy defined in a mapping file. This is based on the observation that no single strategy can be suitable for all different types of stocks. This mapping file should be built based on backtesting results.

Can't download futuOpenD via https://www.futunn.com/download/OpenAPI right now.

Create a data validator to ensure all downloaded data are correct. Use futu openAPI to retrieve trading days and trading day type, then validate the number of entry