Stock Price Prediction Tool

This project provides tools for fetching stock data, preprocessing it, and training a model to predict future stock prices using LSTM neural networks.

Features

• Fetch historical stock data using the yfinance library.
• Store stock data in a PostgreSQL database.
• Preprocess the data and scale it for training.
• Build and train an LSTM model to predict stock prices.

Requirements

• Python 3.x
• yfinance
• psycopg2
• numpy
• pandas
• scikit-learn
• tensorflow
• PostgreSQL

Installation

1. Clone the repository:

   git clone https://github.com/KunshArora912/Stock-Prediction.git
   cd stock-price-prediction-tool

2. Install the required packages:

   pip install yfinance psycopg2 numpy pandas scikit-learn tensorflow

3. Set up your PostgreSQL database and update the db_config dictionary in data_fetch.py with your database credentials. Make sure to change the PostgreSQL password.

Usage

Fetching and Storing Stock Data

1. Run the data_fetch.py script to fetch and store stock data in the PostgreSQL database:

   python data_fetch.py

Preprocessing Data

1. Preprocess the stock data using the functions provided in data_preprocess.py. These functions scale the data and create datasets suitable for training the LSTM model.

Training the Model

1. Run the model_train.py script to train the LSTM model and predict stock prices:

   python model_train.py

Code Explanation

data_fetch.py

• fetch_stock_data(ticker, start, end): Fetches historical stock data for the given ticker between the specified dates.
• store_stock_data(ticker, data): Stores the fetched stock data in the PostgreSQL database.
• fetch_and_store_stock_data(ticker, start, end): Combines the fetch and store operations.

data_preprocess.py

• preprocess_data(data): Scales the 'Close' prices of the stock data.
• create_dataset(data, time_step=60): Creates datasets for training the LSTM model, based on a specified time step.

model_train.py

• build_model(input_shape): Builds and compiles the LSTM model.
• process_stock(ticker, start, end): Fetches, preprocesses, and trains the model on the stock data. Predicts the stock price for the given ticker.

Database Setup

1. Create a PostgreSQL database named stock_data.
2. Create a table named stock_prices with the following structure:

   CREATE TABLE stock_prices (
       ticker VARCHAR(10),
       date DATE,
       open FLOAT,
       high FLOAT,
       low FLOAT,
       close FLOAT,
       volume BIGINT,
       PRIMARY KEY (ticker, date)
   );

Contributing

Contributions are welcome! Please open an issue or submit a pull request for any improvements or bug fixes.