Short Put Neural Network

Contributors

Will Cutchin
Erik Klem
Yash Gollapudi

Description

This project aims to create a RNN neural network that will take finacial information for a given underlying over a given time and provide a delta value that is suggested to short puts at. This neural network will be trained on historical data over a given time and aim to give the highest sensable delta value whilest minimizing chances of being in the money at expiration.

Dependencies

Our project has the following dependencies yfinance, pandas, numpy, math, sklearn (only for splitting data): All of them can be downloaded through pip (or pip3) The pip commands for all are as follows: pip install yfinance, pip install pandas, pip install numpy, math is built in, pip install scikit-learn

Assumptions

Strategy
- Shorting put option
Days To Expiration
- Closest monthly expiration to 45 days DTE (round up if less than 35 days)
Selling
- Selling one put contract at every strike from the 50Δ to 20Δ

📌 Objectives

Create a neural network to find a informed data value for short puts on a given underlying
- Implement a RNN Neural Network From Scratch
  - Clean and Gather Financial Data for Training
  - Determine Inputs and Outputs for The Neural Network
  - Train the Neural Network With Financial Data
- Create Visualization of Results
- Release the Neural Network to Handle Real Time Data

🗓 Task Table

Task	Time required	Assigned to	Current Status	Finished
RNN Neural Net Theory	> 4 Days	Will/Erik/Yash	Done	[ X]
RNN Neural Net Implementation	> 1 Week	Will/Erik/Yash	Done	[ X]
RNN Neural Net Training	> 2 Days	Will/Erik/Yash	Done	[ X]
RNN Neural Net Real Time	> 1 Days	Will/Erik/Yash	Done	[X ]
NN Collect Data	> 3 Days	Will/Yash	Done	[ X]
NN Prepare Data	> 2 hours	Will/Yash	Done	[X ]
NN Object Cache	> 1 hours	Will/Yash	Done	[ X]

webclinic017 / short-put-neural-network Goto Github PK