ARBIMETRA LAB
Exponential Smoothing
Forecasts with simple exponential smoothing, Holt's trend method or Holt-Winters seasonal methods.
What it does
Forecasts with simple exponential smoothing, Holt's trend method or Holt-Winters seasonal methods.
Use it for quick, adaptive forecasts of level, trend and seasonality.
Data you need
- One series
Optional: Seasonal period
Supported data: Time series. Level: Intermediate.
Formula and method
ℓₜ = αyₜ + (1−α)(ℓₜ₋₁ + bₜ₋₁)
Level, trend and seasonal states are updated recursively with smoothing parameters and initial states estimated by maximum likelihood in the ETS framework, which also provides forecast intervals.
Assumptions
- Errors are additive and uncorrelated
Input checks
- Multiplicative seasonality needs positive values
- At least two seasonal cycles
How to read the result
Parameters near 1 adapt quickly and near 0 change slowly.
Worked example
Illustrative run on Box-Jenkins airline passengers. These figures are the actual output of this model on that example, and they say nothing about your own data.
- AIC = 1,123.11, in-sample RMSE = 10.5462, Ljung-Box p = 0.1123.
- Remaining residual autocorrelation is not statistically significant at the 5% level.
Limitations
- Results depend on the sample period
Numerical validation
Validated 16 of 16 automated checks passed, including numerical results, invalid input, missing values, small samples, constant variables, charts and the interpretation text.
| Check | Reference | Result | Abs. error | Tolerance | Source |
|---|---|---|---|---|---|
| SES forecast equals recursion | 431.99580071 | 431.99580071 | 0.0e+0 | 1e-6 | Exponential smoothing recursion |
| Holt forecast equals recursion | 489.99785898 | 489.99785898 | 0.0e+0 | 1e-5 | ETS recursion |
Computed with python 3.13.2, numpy 2.2.5, scipy 1.14.1, pandas 2.3.3, statsmodels 0.14.4, scikit-learn 1.7.0, in your browser. A passing check confirms the listed comparison only. It does not make any result suitable for your data, and it is not a claim of equivalence with other statistical software.
Questions
Which method should I choose?
Use simple smoothing with no trend or seasonality, Holt with a trend, and Holt-Winters with seasonality.
ARBIMETRA LAB is an educational and research tool. Results should be checked by the researcher and are not professional advice.
