Added more unit tests. Added more Python versions. Removed need to se…

…t ENVVAR for example scripts.

.coveragerc

@@ -0,0 +1,7 @@
+[report]
+exclude_lines =
+    # Have to re-enable the standard pragma
+    pragma: no cover
+
+    # Don't complain if tests don't hit defensive assertion code:
+    raise NotImplementedError

.travis.yml

@@ -1,7 +1,10 @@
 language: python
 
 python:
+  - "3.5"
+  - "3.6"
   - "3.7"
+  - "3.8"
 
 before_install:
   - export PYTHONPATH=$PYTHONPATH:$(pwd) 

CHANGELOG.md

@@ -0,0 +1,9 @@
+# 0.1.1
+
+* Removed the need to specify a CSV data directory as an environment variable by adding a default of the current working directory of the executed script
+* Addes CI support for Python 3.5, 3.6 and 3.8 in addition to 3.7
+* Added some unit tests to improve test coverage slightly
+
+# 0.1.0
+
+* Initial relase of QSTrader to PyPI

MANIFEST.in

@@ -1 +1 @@
-include README.md LICENSE
+include README.md LICENSE CHANGELOG.md

README.md

@@ -1,18 +1,19 @@
-![GitHub](https://img.shields.io/github/license/mhallsmoore/qstrader)
-[![Build Status](https://travis-ci.org/mhallsmoore/qstrader.svg?branch=master)](https://travis-ci.org/mhallsmoore/qstrader)
-[![Coverage Status](https://coveralls.io/repos/github/mhallsmoore/qstrader/badge.svg?branch=development)](https://coveralls.io/github/mhallsmoore/qstrader?branch=master)
-[![PyPI](https://img.shields.io/pypi/v/qstrader)](https://pypi.org/project/qstrader)
-[![Python Version](https://img.shields.io/pypi/pyversions/qstrader)](https://pypi.org/project/qstrader)
-
 # QSTrader
 
+| Development   | Details       |
+| ------------- | ------------- |
+| Test Status   | [![Build Status](https://img.shields.io/travis/mhallsmoore/qstrader?label=TravisCI&style=flat-square)](https://travis-ci.org/mhallsmoore/qstrader) [![Coverage Status](https://img.shields.io/coveralls/github/mhallsmoore/qstrader?style=flat-square&label=Coverage)](https://coveralls.io/github/mhallsmoore/qstrader?branch=master) |
+| Version Info  | [![PyPI](https://img.shields.io/pypi/v/qstrader?style=flat-square&label=PyPI&color=blue)](https://pypi.org/project/qstrader) [![PyPI Downloads](https://img.shields.io/pypi/dm/qstrader?style=flat-square&label=PyPI%20Downloads)](https://pypi.org/project/qstrader) |
+| Compatibility | [![Python Version](https://img.shields.io/pypi/pyversions/qstrader?style=flat-square&label=Python%20Versions)](https://pypi.org/project/qstrader) |
+| License       | ![GitHub](https://img.shields.io/github/license/mhallsmoore/qstrader?style=flat-square&label=License) |      
+
 QSTrader is a free Python-based open-source modular schedule-driven backtesting framework for long-only equities and ETF based systematic trading strategies.
 
 QSTrader can be best described as a loosely-coupled collection of modules for carrying out end-to-end backtests with realistic trading mechanics. 
 
 The default modules provide useful functionality for certain types of systematic trading strategies and can be utilised without modification. However the intent of QSTrader is for the users to extend, inherit or fully replace each module in order to provide custom functionality for their own use case.
 
-The software is currently under active development (semantic version 0.x.x) and is provided under a permissive "MIT" license.
+The software is currently under active development and is provided under a permissive "MIT" license.
 
 # Previous Version and Advanced Algorithmic Trading
 
@@ -36,17 +37,9 @@ The QSTrader repository provides some simple example strategies at [/examples](h
 
 Within this quickstart section a classic 60/40 equities/bonds portfolio will be backtested with monthly rebalancing on the last day of the calendar month. 
 
-To get started download the [sixty_forty.py](https://github.com/mhallsmoore/qstrader/blob/master/examples/sixty_forty.py) and place into the directory of your choice.
-
-The 60/40 script makes use of OHLC 'daily bar' data from Yahoo Finance. In particular it requires the [SPY](https://finance.yahoo.com/quote/SPY/history?p=SPY) and [AGG](https://finance.yahoo.com/quote/AGG/history?p=AGG) ETFs data. Download the full history for each and save as CSV files in the directory of your choice.
-
-In order to help QSTrader find your data files it is necessary to set an environment variable called ``QSTRADER_CSV_DATA_DIR``. As an example, on a Linux based system, if you have placed your data files underneath the ``/home/myusername/data`` directory then the following line will set the correct value:
-
-```
-export QSTRADER_CSV_DATA_DIR='/home/myusername/data'
-```
+To get started download the [sixty_forty.py](https://github.com/mhallsmoore/qstrader/blob/master/examples/sixty_forty.py) file and place into the directory of your choice.
 
-Alternatively it is possible to add the above line to your environment configuration (such as within the ``.bashrc`` file) to ensure QSTrader can find data files with any new terminal session.
+The 60/40 script makes use of OHLC 'daily bar' data from Yahoo Finance. In particular it requires the [SPY](https://finance.yahoo.com/quote/SPY/history?p=SPY) and [AGG](https://finance.yahoo.com/quote/AGG/history?p=AGG) ETFs data. Download the full history for each and save as CSV files in same directory as ``sixty_forty.py``.
 
 Assuming that an appropriate Python environment exists and QSTrader has been installed via pip (see **Installation** above), make sure to activate the virtual environment, navigate to the directory with ``sixty_forty.py`` and type:
 
@@ -58,9 +51,9 @@ You will then see some console output as the backtest simulation engine runs thr
 
 ![Image of 60/40 Backtest](https://quantstartmedia.s3.amazonaws.com/images/qstrader_sixty_forty_backtest.png)
 
-You can examine the commented ``sixty_forty.py`` file to see the current QSTrader API.
+You can examine the commented ``sixty_forty.py`` file to see the current QSTrader backtesting API.
 
-If you have any questions about the installation or example usage then please feel free to email [support@quantstart.com](mailto:support@quantstart.com).
+If you have any questions about the installation or example usage then please feel free to email [support@quantstart.com](mailto:support@quantstart.com) or raise an issue [here](https://github.com/mhallsmoore/qstrader/issues).
 
 # Current Features
 

examples/buy_and_hold.py

@@ -23,7 +23,7 @@
 
     # To avoid loading all CSV files in the directory, set the
     # data source to load only those provided symbols
-    csv_dir = os.environ.get('QSTRADER_CSV_DATA_DIR')
+    csv_dir = os.environ.get('QSTRADER_CSV_DATA_DIR', '.')
     data_source = CSVDailyBarDataSource(csv_dir, Equity, csv_symbols=strategy_symbols)
     data_handler = BacktestDataHandler(strategy_universe, data_sources=[data_source])
 

examples/sixty_forty.py

@@ -23,7 +23,7 @@
 
     # To avoid loading all CSV files in the directory, set the
     # data source to load only those provided symbols
-    csv_dir = os.environ.get('QSTRADER_CSV_DATA_DIR')
+    csv_dir = os.environ.get('QSTRADER_CSV_DATA_DIR', '.')
     data_source = CSVDailyBarDataSource(csv_dir, Equity, csv_symbols=strategy_symbols)
     data_handler = BacktestDataHandler(strategy_universe, data_sources=[data_source])
 

examples/sixty_forty_fees.py

@@ -25,7 +25,7 @@
 
     # To avoid loading all CSV files in the directory, set the
     # data source to load only those provided symbols
-    csv_dir = os.environ.get('QSTRADER_CSV_DATA_DIR')
+    csv_dir = os.environ.get('QSTRADER_CSV_DATA_DIR', '.')
     data_source = CSVDailyBarDataSource(csv_dir, Equity, csv_symbols=strategy_symbols)
     data_handler = BacktestDataHandler(strategy_universe, data_sources=[data_source])
 

qstrader/alpha_model/single_signal.py

@@ -43,4 +43,4 @@ def __call__(self, dt):
             The Asset symbol keyed scalar-valued signals.
         """
         assets = self.universe.get_assets(dt)
-        return {asset.symbol: self.signal for asset in assets}
+        return {asset: self.signal for asset in assets}

qstrader/system/rebalance/end_of_month.py

@@ -75,14 +75,3 @@ def _generate_rebalances(self):
             for date in rebalance_dates
         ]
         return rebalance_times
-
-    def output_rebalances(self):
-        """
-        Output the rebalance timestamp list.
-
-        Returns
-        -------
-        `list[pd.Timestamp]`
-            The list of rebalance timestamps.
-        """
-        return self.rebalances

requirements/base.txt

@@ -1,5 +1,5 @@
 Click>=7.0
-matplotlib>=3.1.1
+matplotlib>=3.0.3
 numpy>=1.17.1
 pandas>=0.25.1
 seaborn>=0.9.0

scripts/static_backtest.py

@@ -55,7 +55,7 @@ def obtain_allocations(allocations):
 @click.option('--id', 'strat_id', help='Backtest strategy ID string')
 @click.option('--tearsheet', 'tearsheet', is_flag=True, default=False, help='Whether to display the (blocking) tearsheet plot')
 def cli(start_date, end_date, allocations, strat_title, strat_id, tearsheet):
-    csv_dir = os.environ.get('QSTRADER_CSV_DATA_DIR')
+    csv_dir = os.environ.get('QSTRADER_CSV_DATA_DIR', '.')
 
     start_dt = pd.Timestamp('%s 00:00:00' % start_date, tz=pytz.UTC)
 

setup.py

@@ -5,7 +5,7 @@
 
 setup(
     name="qstrader",
-    version="0.1.0",
+    version="0.1.1",
     description="QSTrader backtesting simulation engine",
     long_description=long_description,
     long_description_content_type="text/markdown",
@@ -15,19 +15,20 @@
     license="MIT",
     classifiers=[
         "License :: OSI Approved :: MIT License",
-	"Programming Language :: Python",
-        "Programming Language :: Python :: 3",        
-	"Programming Language :: Python :: 3.6",
+        "Programming Language :: Python",
+        "Programming Language :: Python :: 3",
+        "Programming Language :: Python :: 3.5",
+        "Programming Language :: Python :: 3.6",
         "Programming Language :: Python :: 3.7",
+        "Programming Language :: Python :: 3.8"
     ],
     packages=find_packages(exclude=("tests",)),
     include_package_data=True,
     install_requires=[
-	"Click==7.1.2",
-	"matplotlib==3.2.1",
+        "Click==7.1.2",
+        "matplotlib==3.0.3",
         "numpy==1.18.4",
-	"pandas==1.0.3",
-	"seaborn==0.10.1"
+        "pandas==1.0.3",
+        "seaborn==0.10.1"
     ]
 )
-

tests/unit/alpha_model/test_fixed_signals.py

@@ -0,0 +1,28 @@
+from unittest.mock import Mock
+
+import pandas as pd
+import pytest
+import pytz
+
+from qstrader.alpha_model.fixed_signals import FixedSignalsAlphaModel
+
+
+@pytest.mark.parametrize(
+    'signals',
+    [
+        ({'EQ:SPY': 0.75, 'EQ:AGG': 0.75, 'EQ:GLD': 0.75}),
+        ({'EQ:SPY': -0.25, 'EQ:AGG': -0.25, 'EQ:GLD': -0.25})
+    ]
+)
+def test_fixed_signals_alpha_model(signals):
+    """
+    Checks that the fixed signals alpha model correctly produces
+    the same signals for each asset in the universe.
+    """
+    universe = Mock()
+    universe.get_assets.return_value = ['EQ:SPY', 'EQ:AGG', 'EQ:GLD']
+
+    alpha = FixedSignalsAlphaModel(universe=universe, signal_weights=signals)
+    dt = pd.Timestamp('2019-01-01 15:00:00', tz=pytz.utc)
+
+    assert alpha(dt) == signals

tests/unit/alpha_model/test_single_signal.py

@@ -0,0 +1,28 @@
+from unittest.mock import Mock
+
+import pandas as pd
+import pytest
+import pytz
+
+from qstrader.alpha_model.single_signal import SingleSignalAlphaModel
+
+
+@pytest.mark.parametrize(
+    'signal,expected_signals',
+    [
+        (0.75, {'EQ:SPY': 0.75, 'EQ:AGG': 0.75, 'EQ:GLD': 0.75}),
+        (-0.25, {'EQ:SPY': -0.25, 'EQ:AGG': -0.25, 'EQ:GLD': -0.25})
+    ]
+)
+def test_single_signal_alpha_model(signal, expected_signals):
+    """
+    Checks that the single signal alpha model correctly produces
+    the same signal for each asset in the universe.
+    """
+    universe = Mock()
+    universe.get_assets.return_value = ['EQ:SPY', 'EQ:AGG', 'EQ:GLD']
+
+    alpha = SingleSignalAlphaModel(universe=universe, signal=signal)
+    dt = pd.Timestamp('2019-01-01 15:00:00', tz=pytz.utc)
+
+    assert alpha(dt) == expected_signals

tests/unit/portcon/test_pcm.py

@@ -16,7 +16,8 @@
     [
         (
             'empty on both sides',
-            {}, [], []),
+            {}, [], []
+        ),
         (
             'partially intersecting set of assets',
             {

tests/unit/system/rebalance/test_buy_and_hold_rebalance.py

@@ -0,0 +1,20 @@
+import pandas as pd
+import pytest
+import pytz
+
+from qstrader.system.rebalance.buy_and_hold import BuyAndHoldRebalance
+
+
+@pytest.mark.parametrize(
+    "start_dt", [('2020-01-01'), ('2020-02-02')]
+)
+def test_buy_and_hold_rebalance(start_dt):
+    """
+    Checks that the buy and hold rebalance sets the
+    appropriate internal attributes.
+    """
+    sd = pd.Timestamp(start_dt, tz=pytz.UTC)
+    reb = BuyAndHoldRebalance(start_dt=sd)
+
+    assert reb.start_dt == sd
+    assert reb.rebalances == [sd]

tests/unit/system/rebalance/test_end_of_month_rebalance.py

@@ -0,0 +1,48 @@
+import pandas as pd
+import pytest
+import pytz
+
+from qstrader.system.rebalance.end_of_month import EndOfMonthRebalance
+
+
+@pytest.mark.parametrize(
+    "start_date,end_date,pre_market,expected_dates,expected_time",
+    [
+        (
+            '2020-03-11', '2020-12-31', False, [
+                '2020-03-31', '2020-04-30', '2020-05-29', '2020-06-30',
+                '2020-07-31', '2020-08-31', '2020-09-30', '2020-10-30',
+                '2020-11-30', '2020-12-31'
+            ], '21:00:00'
+        ),
+        (
+            '2019-12-26', '2020-09-01', True, [
+                '2019-12-31', '2020-01-31', '2020-02-28', '2020-03-31',
+                '2020-04-30', '2020-05-29', '2020-06-30', '2020-07-31',
+                '2020-08-31'
+            ], '14:30:00'
+        )
+    ]
+)
+def test_monthly_rebalance(
+    start_date, end_date, pre_market, expected_dates, expected_time
+):
+    """
+    Checks that the end of month (business day) rebalance provides
+    the correct datetimes for the provided range.
+    """
+    sd = pd.Timestamp(start_date, tz=pytz.UTC)
+    ed = pd.Timestamp(end_date, tz=pytz.UTC)
+
+    reb = EndOfMonthRebalance(
+        start_dt=sd, end_dt=ed, pre_market=pre_market
+    )
+
+    actual_datetimes = reb._generate_rebalances()
+
+    expected_datetimes = [
+        pd.Timestamp('%s %s' % (expected_date, expected_time), tz=pytz.UTC)
+        for expected_date in expected_dates
+    ]
+
+    assert actual_datetimes == expected_datetimes

tests/unit/system/rebalance/test_weekly_rebalance.py

@@ -0,0 +1,64 @@
+import pandas as pd
+import pytest
+import pytz
+
+from qstrader.system.rebalance.weekly import WeeklyRebalance
+
+
+@pytest.mark.parametrize(
+    "start_date,end_date,weekday,pre_market,expected_dates,expected_time",
+    [
+        (
+            '2020-03-11', '2020-05-17', 'MON', False, [
+                '2020-03-16', '2020-03-23', '2020-03-30', '2020-04-06',
+                '2020-04-13', '2020-04-20', '2020-04-27', '2020-05-04',
+                '2020-05-11'
+            ], '21:00:00'
+        ),
+        (
+            '2019-12-26', '2020-02-07', 'WED', True, [
+                '2020-01-01', '2020-01-08', '2020-01-15', '2020-01-22',
+                '2020-01-29', '2020-02-05'
+            ], '14:30:00'
+        )
+    ]
+)
+def test_weekly_rebalance(
+    start_date, end_date, weekday, pre_market, expected_dates, expected_time
+):
+    """
+    Checks that the weekly rebalance provides the correct business
+    datetimes for the provided range.
+    """
+    sd = pd.Timestamp(start_date, tz=pytz.UTC)
+    ed = pd.Timestamp(end_date, tz=pytz.UTC)
+
+    reb = WeeklyRebalance(
+        start_date=sd, end_date=ed, weekday=weekday, pre_market=pre_market
+    )
+
+    actual_datetimes = reb._generate_rebalances()
+
+    expected_datetimes = [
+        pd.Timestamp('%s %s' % (expected_date, expected_time), tz=pytz.UTC)
+        for expected_date in expected_dates
+    ]
+
+    assert actual_datetimes == expected_datetimes
+
+
+def test_check_weekday_raises_value_error():
+    """
+    Checks that initialisation of WeeklyRebalance raises
+    a ValueError if the weekday string is in the incorrect
+    format.
+    """
+    sd = pd.Timestamp('2020-01-01', tz=pytz.UTC)
+    ed = pd.Timestamp('2020-02-01', tz=pytz.UTC)
+    pre_market = True
+    weekday = 'SUN'
+
+    with pytest.raises(ValueError):
+        WeeklyRebalance(
+            start_date=sd, end_date=ed, weekday=weekday, pre_market=pre_market
+        )