Column introduction to time series forecasting—algorithm principles, source code analysis, and project practice

Column link:https://blog.csdn.net/qq_41921826/category_12495091.html

Column content

 All articles provide source code, data sets, and effect visualization

 The article has been on the field content list, daily must-read list, and comprehensive hot list of the entire site many times.

Problems with time series forecasting

 A large number of existing methods do not really predict future values, but only use historical data for verification.

 There is a problem of information leakage when using time series decomposition algorithm:Someone uses emd+lstm to predict time series. Is there any principle problem? - Zhihu

 Welcome to subscribe. When the number of articles reaches a certain number, you will be charged.

 The article is being updated continuously. Complete the blue link to enter the article...

Chapter 1: Essential Basics

Time Series Forecasting—A review of time series forecasting research

Time series forecasting — forecasting data sets (load, wind power, photovoltaic, sales, etc.)

Time Series Forecasting — Understanding Univariate vs. Multivariate

Time Series Forecasting — The problem with time series forecasting: Does it really predict future values?

Time series forecasting — time series data preprocessing

Time series forecasting - time series data feature engineering

Time Series Forecasting - Time Series Data Visualization: Essential for Writing Papers

Time Series Forecasting - Time Series Data Analysis: Essential for Writing Papers

Chapter 2: Statistical Learning Methods

Time series forecasting — ARIMA model principle

Time series forecasting—ARIMA implements single-input single-output load forecasting

Chapter 3: Machine Learning Methods

Time series forecasting - LightGBM implements multi-variable multi-step load forecasting

Time series forecasting - XGBoost implements multi-variable multi-step load forecasting

Time series forecasting - LSSVM implements multi-variable multi-step load forecasting

Time series forecasting - MLP implements multi-variable multi-step load forecasting

Time series forecasting - ELM implements multi-variable multi-step load forecasting

Time series forecasting - CatBoost implements multi-variable multi-step load forecasting

Chapter 4: Deep Learning Methods

Time series forecasting - GRU implements multi-variable multi-step load forecasting (Tensorflow) 

Time series forecasting - LSTM implements univariate rolling wind power forecasting (Tensorflow)

Time series forecasting - LSTM implements multi-variable multi-step load forecasting (Tensorflow)

Time series forecasting - BiLSTM implements multi-variable multi-step photovoltaic forecasting (Tensorflow)

Time series forecasting - BP implements multi-variable multi-step load forecasting (Tensorflow)

Time series forecasting - ARIMA-LSTM implements multi-variable multi-step load forecasting (Tensorflow)

Time series forecasting - CNN-LSTM implements multi-variable multi-step load forecasting (Tensorflow)

Time series forecasting - MLP-LSTM implements multi-variable multi-step load forecasting (Tensorflow)

Chapter 5: Transformer model method

Time Series Forecasting — Transformer Model Principle

Time series prediction - Transformer source code detailed explanation and operation

Time series forecasting — Transformer implements multivariable load forecasting (PyTorch)

Time Series Forecasting — Informer Model Principle

Time Series Forecasting — Detailed explanation and operation of Informer source code

Time series forecasting — Informer implements multivariable load forecasting (PyTorch)

Time Series Forecasting — Autoformer Model Principle

Time Series Forecasting - Detailed explanation and operation of Autoformer source code

Time Series Forecasting — Autoformer implements multi-variable load forecasting

Time series forecasting — CNN-Attention implements multi-variable multi-step load forecasting

Time series forecasting - LSTM-Attention implements multi-variable multi-step load forecasting

Time series prediction - CNN+LSTM+Attention realizes time series prediction

Chapter 6: Time Series Decomposition Method

Time series forecasting—principle of wavelet decomposition model

Time Series Forecasting—VMD Model Principle

Time Series Forecasting—EEMD Model Principle

Time series forecasting — CEEMDAN decomposition model

Time series forecasting - information leakage problem in time series decomposition

Time series forecasting - VMD-LSTM implements single-variable multi-step photovoltaic forecasting (Tensorflow): single variable to multi-variable

Time series forecasting - VMD-LSTM implements single-variable multi-step photovoltaic forecasting (Tensorflow): single variable to multiple single variables

Time series forecasting - CEEMDAN-LSTM realizes multi-variable multi-step load forecasting

Time series forecasting - EMD-VMD-LSTM realizes multi-variable multi-step load forecasting

Chapter 7: Optimization algorithm improvement methods

Time series forecasting - (bat) BAT-LSTM realizes multi-variable multi-step load forecasting

Time series forecasting - (genetic) GA-LSTM realizes multi-variable multi-step load forecasting

Time series forecasting - (Whale) WOA-LSTM realizes multi-variable multi-step load forecasting

Time series forecasting - (Firefly) FA-LSTM realizes multi-variable multi-step load forecasting

Time series forecasting - (Particle Swarm) PSO-LSTM realizes multi-variable multi-step load forecasting

Time series forecasting - (Sparrow Search) SSA-GRU implements multi-variable multi-step load forecasting

Chapter 8: Residual correction and improvement method

Time series forecasting - ARIMA-LSTM realizes multi-variable multi-step load forecasting

Time series forecasting - ARIMA-CNN-LSTM realizes multi-variable multi-step load forecasting

Time series forecasting - ARIMA-WOA-LSTM realizes multi-variable multi-step load forecasting

Time series forecasting - ARIMA-WOA-CNN-LSTM realizes multi-variable multi-step load forecasting

Time series forecasting - ARIMA-LSTM-attention realizes multi-variable multi-step load forecasting

Time series forecasting - MLP-LSTM realizes multi-variable multi-step load forecasting

Chapter 9: Time Series Forecasting Competition Case

Time Series Forecasting — (Competition 1) ‘Teddy Cup’ Data Mining Challenge