Fixed Short put spread ratio issue (#15)

* Minor bug fixes, and code improvements

* added trade numbers to results dataframe

* refactored data.py, filter.py, fixed short put spreads

Author: michaelchu

.gitignore

@@ -33,4 +33,6 @@ MANIFEST
 
 /data/
 /tests/.pytest_cache/
+/strategies/data/
+/strategies/results/
 

.travis.yml

@@ -1,14 +1,21 @@
+env:
+  global:
+    - CC_TEST_REPORTER_ID=9a990fea3fb57063b45010735b989c837fbf4b5da5d4bdfeafb89539e2b61d19
 language: python
 python:
   - "3.6"
+before_script:
+  - curl -L https://codeclimate.com/downloads/test-reporter/test-reporter-latest-linux-amd64 > ./cc-test-reporter
+  - chmod +x ./cc-test-reporter
+  - ./cc-test-reporter before-build
 # command to install dependencies
 install:
   - pip install -r requirements.txt
-  - pip install codecov
+  - pip install coverage
   - pip install pytest pytest-cov
 # command to run tests
 script:
   - pytest --cov=./
 
-after_success:
-  - codecov
+after_script:
+  - ./cc-test-reporter after-build -t coverage.py --debug --exit-code $TRAVIS_TEST_RESULT

README.md

@@ -1,7 +1,7 @@
 [![Downloads](https://pepy.tech/badge/optopsy)](https://pepy.tech/project/optopsy)
-[![Codacy Badge](https://api.codacy.com/project/badge/Grade/2de8f5b3fa2742de93fb60b3a1ae5683)](https://app.codacy.com/app/michaelchu/optopsy?utm_source=github.com&utm_medium=referral&utm_content=michaelchu/optopsy&utm_campaign=badger)
+[![Maintainability](https://api.codeclimate.com/v1/badges/37b11e992a6900d30310/maintainability)](https://codeclimate.com/github/michaelchu/optopsy/maintainability)
 [![Build Status](https://travis-ci.org/michaelchu/optopsy.svg?branch=master)](https://travis-ci.org/michaelchu/optopsy)
-[![codecov](https://codecov.io/gh/michaelchu/optopsy/branch/master/graph/badge.svg)](https://codecov.io/gh/michaelchu/optopsy)
+[![Test Coverage](https://api.codeclimate.com/v1/badges/37b11e992a6900d30310/test_coverage)](https://codeclimate.com/github/michaelchu/optopsy/test_coverage)
 
 # Optopsy
 
@@ -35,23 +35,20 @@ the rapid development of complex options trading strategies.
 * Spread delta
 * Spread price
 
-### Planned Features
-* Indicator Support - Create entry and exit rules based on indicators
-* Optimizer - Allows users to run multiple backtests with different combinations of parameters
-* Option strategy support:
-    * Single Calls/Puts
-    * Vertical Spreads
-    * Iron Condors (Iron Butterflies)
-    * Covered Stock
-    * Combos (Synthetics/Collars)
-    * Diagonal Spreads
-    * Calendar Spreads
-    * Custom Spreads
-    * Strangles
-    * Straddles
+### Option strategy support
+* Single Calls/Puts
+* Vertical Spreads
+* (Coming Soon) Iron Condors (Iron Butterflies)
+* (Coming Soon) Covered Stock
+* (Coming Soon) Combos (Synthetics/Collars)
+* (Coming Soon) Diagonal Spreads
+* (Coming Soon) Calendar Spreads
+* (Coming Soon) Custom Spreads
+* (Coming Soon) Strangles
+* (Coming Soon) Straddles
 
 ### Dependencies
-You will need Python 3.6.x. It is recommended to install [Miniconda3](https://conda.io/miniconda.html). See [requirements.txt](https://github.com/michaelchu/optopsy/blob/master/requirements.txt) for full details.
+You will need Python 3.6.x and Pandas 0.23.1 or newer. It is recommended to install [Miniconda3](https://conda.io/miniconda.html). See [requirements.txt](https://github.com/michaelchu/optopsy/blob/master/requirements.txt) for full details.
 
 ### Installation
 ```
@@ -69,7 +66,6 @@ In order to use it, you will need to define the struct variable to map the colum
 First we import the library and other nessesary libaries:
 ```python
 import optopsy as op
-import os
 from datetime import datetime
 ```
 
@@ -112,13 +108,9 @@ def run_strategy():
     start = datetime(2016, 1, 1)
     end = datetime(2016, 12, 31)
 
-    # create the option spread that matches the entry filters
+    # create the option spreads that matches the entry filters
     trades = op.strategies.short_call_spread(data, start, end, filters)
 
-    # we get a dataframe of our orders based on the entry filter rules, let's export
-    # this to csv file for reference
-    trades.to_csv('./strategies/results/trades.csv')
-
     # call the run method with our data, option spreads and filters to run the backtest
     backtest = op.run(data, trades, filters)
 

optopsy/__init__.py

@@ -1,5 +1,4 @@
 from .data import get as get
-from .data import gets as gets
 from .enums import *
 from .backtest import run
 import optopsy.option_strategies as strategies

optopsy/backtest.py

@@ -22,73 +22,60 @@
 from .option_queries import opt_type
 from .statistics import *
 
-pd.set_option('display.expand_frame_repr', False)
-
-sort_by = [
-    'underlying_symbol',
-    'quote_date',
-    'option_type',
-    'expiration',
-    'strike'
-]
+pd.set_option("display.expand_frame_repr", False)
 
-on = [
-    'underlying_symbol',
-    'option_type',
-    'expiration',
-    'strike'
-]
+
+on = ["underlying_symbol", "option_type", "expiration", "strike"]
 
 default_entry_filters = {
     "std_expr": False,
     "contract_size": 10,
     "entry_dte": (27, 30, 31),
-    "exit_dte": None
+    "exit_dte": None,
 }
 
 output_cols = {
-    'quote_date_entry': 'entry_date',
-    'quote_date_exit': 'exit_date',
-    'delta_entry': 'entry_delta',
-    'underlying_price_entry': 'entry_stk_price',
-    'underlying_price_exit': 'exit_stk_price',
-    'dte_entry': 'DTE'
+    "quote_date_entry": "entry_date",
+    "quote_date_exit": "exit_date",
+    "delta_entry": "entry_delta",
+    "underlying_price_entry": "entry_stk_price",
+    "underlying_price_exit": "exit_stk_price",
+    "dte_entry": "dte",
 }
 
 output_format = [
-    'entry_date',
-    'exit_date',
-    'expiration',
-    'DTE',
-    'ratio',
-    'contracts',
-    'option_type',
-    'strike',
-    'entry_delta',
-    'entry_stk_price',
-    'exit_stk_price',
-    'entry_opt_price',
-    'exit_opt_price',
-    'entry_price',
-    'exit_price',
-    'profit'
+    "entry_date",
+    "exit_date",
+    "expiration",
+    "underlying_symbol",
+    "dte",
+    "ratio",
+    "contracts",
+    "option_type",
+    "strike",
+    "entry_delta",
+    "entry_stk_price",
+    "exit_stk_price",
+    "entry_opt_price",
+    "exit_opt_price",
+    "entry_price",
+    "exit_price",
+    "profit",
 ]
 
 
 def _create_legs(data, leg):
-    return (
-        data
-        .pipe(opt_type, option_type=leg[0])
-        .assign(ratio=leg[1])
-    )
+    return data.pipe(opt_type, option_type=leg[0]).assign(ratio=leg[1])
 
 
 def _apply_filters(legs, filters):
     if not filters:
         return legs
     else:
-        return [reduce(lambda l, f: getattr(fil, f)(l, filters[f], idx), filters, leg)
-                for idx, leg in enumerate(legs)]
+        return [
+            reduce(lambda l, f: getattr(fil, f)(l, filters[f], idx), filters, leg)
+            for idx, leg in enumerate(legs)
+        ]
 
 
 def _filter_data(data, filters):
@@ -101,37 +88,32 @@ def create_spread(data, leg_structs, filters):
 
     # merge and apply leg filters to create spread
     filters = {**default_entry_filters, **filters}
-    entry_filters = {f: filters[f]
-                     for f in filters if (not f.startswith('entry_spread') and
-                                          not f.startswith('exit_'))}
+    entry_filters = {
+        f: filters[f]
+        for f in filters
+        if (not f.startswith("entry_spread") and not f.startswith("exit_"))
+    }
     spread = _filter_data(legs, entry_filters)
 
     # apply spread level filters to spread
-    spread_filters = {f: filters[f]
-                      for f in filters if f.startswith('entry_spread')}
-    return _filter_data(spread, spread_filters).sort_values(sort_by)
+    spread_filters = {f: filters[f] for f in filters if f.startswith("entry_spread")}
+    return _filter_data(spread, spread_filters)
 
 
 # this is the main function that runs the backtest engine
-def run(data, trades, filters, init_balance=10000, mode='midpoint'):
-    trades = trades if isinstance(trades, list) else [trades]
-
-    # merge trades from multiple underlying symbols if applicable
-    all_trades = pd.concat(trades).sort_values(sort_by)
-
+def run(data, trades, filters, init_balance=10000, mode="midpoint"):
     # for each option to be traded, determine the historical price action
     filters = {**default_entry_filters, **filters}
-    exit_filters = {f: filters[f] for f in filters if f.startswith('exit_')}
+    exit_filters = {f: filters[f] for f in filters if f.startswith("exit_")}
     res = (
-        pd
-        .merge(all_trades, data, on=on, suffixes=('_entry', '_exit'))
+        pd.merge(trades, data, on=on, suffixes=("_entry", "_exit"))
         .pipe(_filter_data, exit_filters)
         .pipe(calc_entry_px, mode)
         .pipe(calc_exit_px, mode)
         .pipe(calc_pnl)
-        # .pipe(calc_running_balance, init_balance)
         .rename(columns=output_cols)
-        .sort_values(['entry_date', 'expiration', 'underlying_symbol', 'strike'])
+        .sort_values(["entry_date", "expiration", "underlying_symbol", "strike"])
+        .pipe(assign_trade_num)
     )
 
     return calc_total_profit(res), res[output_format]