How (not) to use Machine Learning for time series forecasting: Avoiding the pitfalls

 By KDnuggets - 2020-12-07

Description

We outline some of the common pitfalls of machine learning for time series forecasting, with a look at time delayed predictions, autocorrelations, stationarity, accuracy metrics, and more.

Summary

  • Time series forecasting is an important area of machine learning.
  • However, while the time component adds additional information, it also makes time series problems more difficult to handle compared to many other prediction tasks.
  • Plotting the cross-correlation between the predicted and real value (below figure), we see a clear peak at a time lag of 1 day, indicating that the model simply uses the previous value as the prediction for the future.
  • Defining the model to predict the difference in values between time steps rather than the value itself, is a much stronger test of the models predictive powers.

 

Topics

  1. Machine_Learning (0.2)
  2. Backend (0.17)
  3. Database (0.09)

Similar Articles

Random Forest for Time Series Forecasting

By Machine Learning Mastery - 2020-11-01

Random Forest is a popular and effective ensemble machine learning algorithm. It is widely used for classification and regression predictive modeling problems with structured (tabular) data sets, e.g. ...

LSTM for time series prediction

By KDnuggets - 2021-01-23

Learn how to develop a LSTM neural network with PyTorch on trading data to predict future prices by mimicking actual values of the time series data.

Time-Series Forecasting with Google BigQuery ML

By Medium - 2021-02-16

If you have worked with any kind of forecasting models, you will know how laborious it can be at times especially when trying to predict multiple variables. From identifying if a time-series is…

Facebook’s Prophet + Deep Learning = NeuralProphet

By Medium - 2020-12-10

While learning about time series forecasting, sooner or later you will encounter the vastly popular Prophet model, developed by Facebook. It gained lots of popularity due to the fact that it provides…

Creating synthetic time series data

By Medium - 2021-02-22

We synthetic versions of a time-series dataset, visualize and analyze the results, and discuss several use cases for synthetic time series data.