S-P-500-Index-Price-Anomaly-Detection-in-Keras

S&P 500 Index Price Anomaly Detection in Time Series Data with Keras

Project Objective:

• Design and Build an LSTM Autoencoder in Keras
• Detect anomalies (sudden price changes) in the S&P 500 index
• Data range: 1986 - 2020

The S&P 500 is a stock performance indicator for the Top 500 Companies listed on the stock exchange in the United States. It is considered as one of the best representations of the United States stock market.

Source of Dataset: Yahoo Finance or Download your dataset at: https://www.marketwatch.com/investing/index/spx/download-data

The dataset is a CSV format with:

1. Daily timestamp
2. Daily Open and Close price
3. Daily High and Low price
4. Daily Volume

The Project will follow the below steps:

1. Import Libraries
2. Load and Inspect the S&P 500 Index Data
3. Data Preprocessing
4. Temporalizing Data and Creating Training and Test Splits
5. Build an LSTM Autoencoder
6. Train the Autoencoder
7. Plot Metrics and Evaluate the Model
8. Detect Anomalies in the S&P 500 Index Data
9. Visualize the Anomalies based on threshold

The required libraries:

• Numpy
• Tensorflow
• Keras
• Pandas
• Seaborn
• Matplotlib
• Plotly
• Scikit Learn