Distinguish between backtest, paper and live modes

My algorithm requires to store internal state in order to functional correctly. I planned to use ObjectStore for that purpose.

However, same ObjectStore shared between all backtests, live and paper modes. Obviously each algorithm session requires to have its own state and I definitely don't want to corrupt state of my live trading. So I need somehow to distinguish between backtest, paper and live modes - so I could store different keys in ObjectStore and this way to maintain separation.

Moreover, my broker is IBKR, and I intend to use it both for live and paper trading (it supports paper trading). So I need to know whether I'm running in live or paper mode.

I'm aware about “LiveMode” property but its boolean - so using it I can know only whether I'm in backtest or not - but that doesn't help to determine whether its paper or live modes. Also, for a question why I not using QC for paper trading - its because I want my paper trading to be close as possible to live, so I prefer to use IBKR for both, to accurately simulate fees, margins, etc.

from AlgorithmImports import *
class CustomAlgorithm(QCAlgorithm):
def initialize(self):
self.set_start_date(2021, 1, 1)
self.set_cash(100000)
self.add_equity("SPY", Resolution.DAILY)
self.spy_close_object_store_key = "spy_close_"
# Determine the trading environment mode
object_store_key_suffix = self.construct_object_store_key()
self.spy_close_object_store_key += object_store_key_suffix
# Check for existing data and load or fetch/store as needed
if self.object_store.contains_key(self.spy_close_object_store_key):
self.load_object_store_data()
else:
self.fetch_and_store_initial_data()
def construct_object_store_key(self) -> str:
# Construct a unique key based on trading mode and environment
mode = self.get_environment_mode()
if mode == AlgorithmMode.BACKTESTING:
return "backtest"
elif mode == AlgorithmMode.LIVE and self.get_brokerage_environment() == BrokerageEnvironment.PAPER:
return "live_paper"
elif mode == AlgorithmMode.LIVE:
return "live_real"
return "other"
def load_object_store_data(self):
self.debug(f"Loading data from object store key: {self.spy_close_object_store_key}")
def fetch_and_store_initial_data(self):
self.debug(f"Fetching and storing initial data for key: {self.spy_close_object_store_key}")
def on_data(self, slice: Slice):
pass
def get_brokerage_environment(self):
# This method should determine and return the brokerage environment
# Implement the method logic as per your brokerage setup requirements
pass
def get_environment_mode(self):
# Stub: Return the current mode
return AlgorithmMode.LIVE # example, replace with actual logic

+ Expand

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.

One possible way to know whether I run in live or paper in IBKR, is by account id (live starts with “U” and paper starts with “D”). However, I can't find a way to locate current broker account ID. Does such API exists?

Mia Alissi INVESTOR

22 days ago

To manage different algorithm modes effectively using the ObjectStore in QuantConnect, address the following:

Determine Mode and Environment: Use enums AlgorithmMode and BrokerageEnvironment.
Construct Unique Keys: Create unique keys for storing data in each mode.
Correct Code Implementations: Ensure method names are correctly defined and used.

Here's an example of how you might implement this in your algorithm:

from AlgorithmImports import *
class CustomAlgorithm(QCAlgorithm):
    def initialize(self):
        self.set_start_date(2021, 1, 1)
        self.set_cash(100000)
        self.add_equity("SPY", Resolution.DAILY)
        self.spy_close_object_store_key = "spy_close_"
        # Determine the trading environment mode
        object_store_key_suffix = self.construct_object_store_key()
        self.spy_close_object_store_key += object_store_key_suffix
        # Check for existing data and load or fetch/store as needed
        if self.object_store.contains_key(self.spy_close_object_store_key):
            self.load_object_store_data()
        else:
            self.fetch_and_store_initial_data()
    def construct_object_store_key(self) -> str:
        # Construct a unique key based on trading mode and environment
        mode = self.get_environment_mode()
        if mode == AlgorithmMode.BACKTESTING:
            return "backtest"
        elif mode == AlgorithmMode.LIVE and self.get_brokerage_environment() == BrokerageEnvironment.PAPER:
            return "live_paper"
        elif mode == AlgorithmMode.LIVE:
            return "live_real"
        return "other"
    def load_object_store_data(self):
        self.debug(f"Loading data from object store key: {self.spy_close_object_store_key}")
    def fetch_and_store_initial_data(self):
        self.debug(f"Fetching and storing initial data for key: {self.spy_close_object_store_key}")
    def on_data(self, slice: Slice):
        pass
    def get_brokerage_environment(self):
        # This method should determine and return the brokerage environment
        # Implement the method logic as per your brokerage setup requirements
        pass
    def get_environment_mode(self):
        # Stub: Return the current mode
        return AlgorithmMode.LIVE  # example, replace with actual logic
+ Expand
- Collapse

Key Points:

Environment Mode Checking: Ensure get_environment_mode and get_brokerage_environment methods are correctly implemented to ascertain the current mode and brokerage environment.
Unique Key Construction: The construct_object_store_key builds a key based on detected mode.
Data Handling: Manage data either by loading existing data or fetching new data to store.

Sources: Object Store - QuantConnect.com, Object Store - QuantConnect.com, Object Store Example Algorithm, Object Store - QuantConnect.com, Brokerage Model Algorithm

Upvote

Evgeni Aizikovich INVESTOR

Well, if I would know how to implement “get_brokerage_environment” or “get_environment_mode” - I won't be asking this question in the first place…

Louis Szeto INVESTOR

QuantConnect | 17 days ago

Hi Evgeni

We cannot access that within the algorithm, but you can use the project ID to create a new folder in the object store:

self.object_store.save(f"{self.project_id}/<YOUR_STATE_NAME>", object)

Best
Louis

Platform

Radically Open-Source Algorithmic Trading Engine

Join Our Discord Channel

Quarterly Open-Source Trading Competition

Draft Discussions

Bookmarked Discussions

SEARCH DISCUSSIONS

364,900 Quants.

VOTE FOR UPCOMING FEATURES

JOIN OUR Community MAILING LIST

Key Points:

IN THIS RESEARCH

PARTICIPANTS

Actions

Join QuantConnect for Free

Platform

SIGN IN

Radically Open-Source Algorithmic Trading Engine

Join Our Discord Channel

Quarterly Open-Source Trading Competition

Draft Discussions

Bookmarked Discussions

SEARCH DISCUSSIONS

364,900 Quants.

VOTE FOR UPCOMING FEATURES

JOIN OUR Community MAILING LIST

Key Points:

IN THIS RESEARCH

PARTICIPANTS

SHARE RESEARCH

SHARE DISCUSSION

SHARE ARTICLE

SHARE

Actions

Join QuantConnect for Free