Pre-Earnings Straddle Option Trader

Description

• Python Trading Algorithm to Identify Profitable Options Strategies Before Earnings Reports.
• Leverages the increased volatility preceding earnings announcements. Buys straddles at predicted strike prices a week before earnings date and sells right before earnings.

File Overview

For Current Day Trading

• get_stocks.py: Returns a list of stocks with earnings reports coming out next week. Input: Current Date, Output: stock_list.csv
• get_options.py: Utilizes decision logic to create a list of suggested straddles to buy based off given company names. Input: stock_list.csv, Output: options_data.xlsx
• decision_making.py: Utilizes a gradient boosting algorithm to sort out unprofitable trades and returns predicted profitable trades. Input: data.xlsx, options_data.xlsx, Output: predicted_options_data.xlsx

Backtesting

• get_stocks_backtest.py: Returns a list of stocks with earnings reports coming out next week. Input: backtest_date, Output: stock_list_backtest.csv
• get_options_backtest.py: Utilizes the same decision logic to create a list of suggested straddles and calculates the return of each straddle trade. Input: stock_list_backtest.csv, Output: options_data_backtest.xlsx
• decision_making_backtest.py: Trains multiple models and determines the model with the best accuracy. Repredicts on the data and returns all suggested options trades. Input: data.xlsx, Output: final_stock_list_backtest.xlsx

Data

• data.xlsx: Cumulative data generated from the backtesting files for training the decision tree model
• final_stock_list_backtest.xlsx: Sample backtested generated stock suggestions over the past 2 months with their individual gains/losses. Showcases an average return of 40% on each trade suggestion.

File Hierarchy Rationale

• Dividing get_stocks and get_options allows for tweaking the decision making algorithm for straddles without waiting to re-retrieve the stock data
• Current suggestions and backtesting are divided because of the differences in API usage and overall code

Requirements

Libraries

pip install the following libraries

• requests
• pandas
• datetime
• yfinance
• pandas_market_calendars
• sklearn
• matplotlib
• seaborn
• numpy

APIs

Enter API Keys

• Orats
• Finnhub
• Alpha Vantage

How To Run

• Current Suggestions: run get_stocks.py -> get_options.py -> decision_making.py. Change variables and logic as needed.
• Backtesting: run get_stocks_backtest.py -> get_options_backtest.py -> decision_making_backtest.py. Change variables and logic as needed.