BackTest: Daylight Savings Fix

lumibot/backtesting/backtesting_broker.py

@@ -4,7 +4,7 @@
 from decimal import Decimal
 from functools import wraps
 
-import pandas as pd
+import pytz
 
 from lumibot.brokers import Broker
 from lumibot.data_sources import DataSourceBacktesting
@@ -88,13 +88,19 @@ def get_historical_account_value(self):
     def _update_datetime(self, update_dt, cash=None, portfolio_value=None):
         """Works with either timedelta or datetime input
         and updates the datetime of the broker"""
+        tz = self.datetime.tzinfo
+        is_pytz = isinstance(tz, (pytz.tzinfo.StaticTzInfo, pytz.tzinfo.DstTzInfo))
 
         if isinstance(update_dt, timedelta):
             new_datetime = self.datetime + update_dt
         elif isinstance(update_dt, int) or isinstance(update_dt, float):
             new_datetime = self.datetime + timedelta(seconds=update_dt)
         else:
             new_datetime = update_dt
+
+        # This is needed to handle Daylight Savings Time changes
+        new_datetime = tz.normalize(new_datetime) if is_pytz else new_datetime
+
         self.data_source._update_datetime(new_datetime, cash=cash, portfolio_value=portfolio_value)
         if self.option_source:
             self.option_source._update_datetime(new_datetime, cash=cash, portfolio_value=portfolio_value)

lumibot/data_sources/interactive_brokers_rest_data.py

@@ -1,8 +1,10 @@
 import logging
 from termcolor import colored
-from ..entities import Asset, Bars
 
+from lumibot import LUMIBOT_DEFAULT_PYTZ
+from ..entities import Asset, Bars
 from .data_source import DataSource
+
 import subprocess
 import os
 import time
@@ -813,7 +815,7 @@ def get_historical_prices(
         # Convert timestamp to datetime and set as index
         df["timestamp"] = pd.to_datetime(df["timestamp"], unit="ms")
         df["timestamp"] = (
-            df["timestamp"].dt.tz_localize("UTC").dt.tz_convert("America/New_York")
+            df["timestamp"].dt.tz_localize("UTC").dt.tz_convert(LUMIBOT_DEFAULT_PYTZ)
         )
         df.set_index("timestamp", inplace=True)
 
@@ -1078,4 +1080,4 @@ def get_quote(self, asset, quote=None, exchange=None):
         else:
             result["ask"] = None
 
-        return result
+        return result

lumibot/data_sources/tradier_data.py

@@ -3,8 +3,8 @@
 from datetime import datetime, date, timedelta
 
 import pandas as pd
-import pytz
 
+from lumibot import LUMIBOT_DEFAULT_PYTZ
 from lumibot.entities import Asset, Bars
 from lumibot.tools.helpers import create_options_symbol, parse_timestep_qty_and_unit, get_trading_days
 from lumiwealth_tradier import Tradier
@@ -189,7 +189,7 @@ def get_historical_prices(
         end_date = datetime.now()
 
         # Use pytz to get the US/Eastern timezone
-        eastern = pytz.timezone("US/Eastern")
+        eastern = LUMIBOT_DEFAULT_PYTZ
 
         # Convert datetime object to US/Eastern timezone
         end_date = end_date.astimezone(eastern)
@@ -242,7 +242,7 @@ def get_historical_prices(
 
         # if type of index is date, convert it to timestamp with timezone info of "America/New_York"
         if isinstance(df.index[0], date):
-            df.index = pd.to_datetime(df.index).tz_localize("America/New_York")
+            df.index = pd.to_datetime(df.index).tz_localize(LUMIBOT_DEFAULT_PYTZ)
 
         # Convert the dataframe to a Bars object
         bars = Bars(df, self.SOURCE, asset, raw=df, quote=quote)

lumibot/strategies/_strategy.py

@@ -21,6 +21,7 @@
 from sqlalchemy import create_engine, inspect, text
 
 import pandas as pd
+from lumibot import LUMIBOT_DEFAULT_PYTZ
 from ..backtesting import BacktestingBroker, PolygonDataBacktesting, ThetaDataBacktesting
 from ..entities import Asset, Position, Order
 from ..tools import (
@@ -1791,7 +1792,7 @@ def send_account_summary_to_discord(self):
         cash = self.get_cash()
 
         # # Get the datetime
-        now = pd.Timestamp(datetime.datetime.now()).tz_localize("America/New_York")
+        now = pd.Timestamp(datetime.datetime.now()).tz_localize(LUMIBOT_DEFAULT_PYTZ)
 
         # Get the returns
         returns_text, stats_df = self.calculate_returns()
@@ -1820,7 +1821,7 @@ def get_stats_from_database(self, stats_table_name, retries=5, delay=5):
                     self.logger.info(f"Table {stats_table_name} does not exist. Creating it now.")
 
                     # Get the current time in New York
-                    ny_tz = pytz.timezone("America/New_York")
+                    ny_tz = LUMIBOT_DEFAULT_PYTZ
                     now = datetime.datetime.now(ny_tz)
 
                     # Create an empty stats dataframe
@@ -1884,7 +1885,7 @@ def backup_variables_to_db(self):
             self.db_engine = create_engine(self.db_connection_str)
 
         # Get the current time in New York
-        ny_tz = pytz.timezone("America/New_York")
+        ny_tz = LUMIBOT_DEFAULT_PYTZ
         now = datetime.datetime.now(ny_tz)
 
         if not inspect(self.db_engine).has_table(self.backup_table_name):
@@ -2008,7 +2009,7 @@ def calculate_returns(self):
         # Calculate the return over the past 24 hours, 7 days, and 30 days using the stats dataframe
 
         # Get the current time in New York
-        ny_tz = pytz.timezone("America/New_York")
+        ny_tz = LUMIBOT_DEFAULT_PYTZ
 
         # Get the datetime
         now = datetime.datetime.now(ny_tz)
@@ -2025,11 +2026,11 @@ def calculate_returns(self):
         # Check if the datetime column is timezone-aware
         if stats_df['datetime'].dt.tz is None:
             # If the datetime is timezone-naive, directly localize it to "America/New_York"
-            stats_df["datetime"] = stats_df["datetime"].dt.tz_localize("America/New_York", ambiguous='infer')
+            stats_df["datetime"] = stats_df["datetime"].dt.tz_localize(LUMIBOT_DEFAULT_PYTZ, ambiguous='infer')
         else:
             # If the datetime is already timezone-aware, first remove timezone and then localize
             stats_df["datetime"] = stats_df["datetime"].dt.tz_localize(None)
-            stats_df["datetime"] = stats_df["datetime"].dt.tz_localize("America/New_York", ambiguous='infer')
+            stats_df["datetime"] = stats_df["datetime"].dt.tz_localize(LUMIBOT_DEFAULT_PYTZ, ambiguous='infer')
 
         # Get the stats
         stats_new = pd.DataFrame(
@@ -2049,7 +2050,7 @@ def calculate_returns(self):
         stats_df = pd.concat([stats_df, stats_new])
 
         # # Convert the datetime column to eastern time
-        stats_df["datetime"] = stats_df["datetime"].dt.tz_convert("America/New_York")
+        stats_df["datetime"] = stats_df["datetime"].dt.tz_convert(LUMIBOT_DEFAULT_PYTZ)
 
         # Remove any duplicate rows
         stats_df = stats_df[~stats_df["datetime"].duplicated(keep="last")]
@@ -2160,4 +2161,4 @@ def calculate_returns(self):
             return results_text, stats_df
 
         else:
-            return "Not enough data to calculate returns", stats_df
+            return "Not enough data to calculate returns", stats_df

lumibot/tools/thetadata_helper.py

@@ -8,7 +8,7 @@
 import pandas as pd
 import pandas_market_calendars as mcal
 import requests
-from lumibot import LUMIBOT_CACHE_FOLDER
+from lumibot import LUMIBOT_CACHE_FOLDER, LUMIBOT_DEFAULT_PYTZ
 from lumibot.entities import Asset
 from thetadata import ThetaClient
 from tqdm import tqdm
@@ -295,7 +295,7 @@ def update_df(df_all, result):
                 ],
             }
     """
-    ny_tz = pytz.timezone('America/New_York')
+    ny_tz = LUMIBOT_DEFAULT_PYTZ
     df = pd.DataFrame(result)
     if not df.empty:
         if "datetime" not in df.index.names: