Time series analysis comprises methods for analyzing time series data in order to extract meaningful statistics and other characteristics of the data. Time series forecasting is the use of a model to predict future values based on previously observed values.
Time series are widely used for non-stationary data, like economic, weather, stock price, and retail sales in this post. In this repository contains different methods for forecasting time series, inclusing:
- basic techniques (Naive method, simple average, simple moding average)
- exponential smoothing techniques (Simple exponential smoothing, Holt's method, Holt Winter's method)
- auto regressive methods (ARIMA, SARIMA)
- FB Prophet
- deep learning techniques (N-BEATS, N-HiTS, PatchTST, TimesNet, TimeGPT, TiDE)
Quick table to chose what is the better technique for forecasting your data:
