Stockr is an end to end ML and DL Powered App that lets you predict/forecast a stock.

Fetches Data from yfinance API

Preprocess and Scales Data.

Feature Engineering : Appends additional useful indicators and features to the data.

Model training : Trains Models like ARIMA, SARIMAX, LSTM, Prophet, etc. on the stock data.

Forecast & Predict : Makes forecast and predictions on the historical engineered data.

Reiterate : Retrains and improves on further iterations.

Stockr Uses yfinance api to extract stock data. Currently, Stockr supports Apple, Microsoft, Amazon, Google, Tesla Stocks data.

Collected data is then cleaned, preprocessed, scaled using Scikit Learn's train test split and Standard Scaler.

We then extract more meaningful features and Indicators from the preprocessed data.

We can simply add features like day of week, quarter, Month, etc.

These Features Include : Rolling Window Features, Lag Features, Exponential Moving Average, Moving Average Convergence Divergence, Relative Strength Index, Bollinger Bands, Volume Based Features.