In this chapter, we introduced the use of a weighted average of past observations for forecast time series data. We started with a simplistic and naive forecasting approach with the moving average function. Although this function is limited to short-term forecasts and can only handle time series with no seasonal and trend components, it provides context for exponential smoothing functions. The exponential smoothing family of forecasting models is based on the use of different smoothing parameters, that is , , and , for modeling the main components of time series data—level, trend, and seasonal, respectively. The main advantages of exponential smoothing functions are their simplicity, they're cheap for computing, and their modularity, which allows them to handle different types of time series data, such as linear and exponential trends and seasonal components...
- Tech Categories
- Best Sellers
- New Releases
- Books
- Videos
- Audiobooks
Tech Categories Popular Audiobooks
- Articles
- Newsletters
- Free Learning
You're reading from Hands-On Time Series Analysis with R
Rami Krispin is a data scientist at a major Silicon Valley company, where he focuses on time series analysis and forecasting. In his free time, he also develops open source tools and is the author of several R packages, including the TSstudio package for time series analysis and forecasting applications. Rami holds an MA in Applied Economics and an MS in actuarial mathematics from the University of MichiganAnn Arbor.
Read more about Rami Krispin
Unlock this book and the full library FREE for 7 days
Author (1)
Rami Krispin is a data scientist at a major Silicon Valley company, where he focuses on time series analysis and forecasting. In his free time, he also develops open source tools and is the author of several R packages, including the TSstudio package for time series analysis and forecasting applications. Rami holds an MA in Applied Economics and an MS in actuarial mathematics from the University of MichiganAnn Arbor.
Read more about Rami Krispin