I am bit new to time series modelling, currently I am trying to understand some basics. What is the difference between smoothing and decomposition in time series. I have gone through many materials, I feel both are same. Please explain in simple way.


2 Answers 2


To start with I will list 7 forecasting methods in time series.

  1. Naive Approach
  2. Simple Average
  3. Moving Average
  4. Simple Exponential Smoothing
  5. Holt’s Linear Trend method
  6. Holt-Winters Method
  7. ARIMA

The first 4 methods try to make the rough edges of time series data smooth so as to correctly forecast the data.

Image(s) to describe smoothing techniques in time series

In the above image blue color shows the trend with true time series data while the red color shows the smoothed series.

The last 3 methods try to break down time series data into its various component such as trend, season, cycle and residual (remnant)

Image(s) to describe decomposition in time series

In the above image, I show what to expect in the decomposition of time series data


It's different.
Smoothing like moving average outputs data smoothed.

Decomposition like seasonal decomposition outputs separating seasonality, trend, level, resid.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.