This project is intended to apply time series forecasting algorithms by using Facebook Stock Price as an example. It is by no means a very practical method to help making decisions for investments. In fact, it is strictly not advisable to use any of this methods as a way to do so.
Models used involved: ARIMA
, SARIMAX
, Prophet
, and LSTM
. LSTM
seems to perform very well when predicting one day into the future using 10 previous time steps. It was able to capture the trends of rising and dropping in this case. Further explanation is available in the Web Application.
Thank you for checking out!
The Web App is accessible here which you can directly see all the visualizations made.