TimeSeriesPrediction

Overview of Time Series Forecasting from Statistical to Recent ML Approaches

These notebooks are supplement material for the blog here: Overview of Time Series Forecasting from Statistical to Recent ML Approaches

Topics and corresponding notebooks:

• Introduction to TS (Time_Series_Intro_ES_ARIMA.ipynb)
• Decompose (Time_Series_FFT.ipynb)
  • Gen Synthic
  • Decompose FFT
• Naive approaches (Time_Series_Intro_ES_ARIMA.ipynb.ipynb)
• Statistical (Time_Series_Intro_ES_ARIMA.ipynb.ipynb)
  • Smoothing techniques (MA, ES)
  • ARIMA
  • State Space (Time_Series_StateSpace.ipynb)
• ML (Time_Series_ML-LR_XGBoost.ipynb)
  • Linear Regression
  • Decision Tree (XGBoost)
• DL (Time_Series_DL_LSTM_CNN.ipynb)
  • LSTM, CNN + LSTM
  • TCN (Time_Series_DL_TCN_LSTNet.ipynb)
  • LSTNet
  • TFT (Time_Series_DL_TFT_N-BEATS.ipynb)
  • N-BEATS
• Commercial: (Time_Series_FBProphet_DeepAR.ipynb)
  • Facebook Prophet
  • Amazon DeepAR