Abstract
In a recent study, Bergmeir, Hyndman and Benítez (2016) successfully employed a bootstrap aggregation (bagging) technique for improving the performance of exponential smoothing. Each series is Box-Cox transformed, and decomposed by Seasonal and Trend decomposition using Loess (STL); then bootstrapping is applied on the remainder series before the trend and seasonality are added back, and the transformation reversed to create bootstrapped versions of the series. Subsequently, they apply automatic exponential smoothing on the original series and the bootstrapped versions of the series, with the final forecast being the equal-weight combination across all forecasts. In this study we attempt to address the question: why does bagging for time series forecasting work? We assume three sources of uncertainty (model uncertainty, data uncertainty, and parameter uncertainty) and we separately explore the benefits of bagging for time series forecasting for each one of them. Our analysis considers 4004 time series (from the M- and M3-competitions) and two families of models. The results show that the benefits of bagging predominantly originate from the model uncertainty: the fact that different models might be selected as optimal for the bootstrapped series. As such, a suitable weighted combination of the most suitable models should be preferred to selecting a single model.
| Original language | English |
|---|---|
| Pages (from-to) | 545-554 |
| Number of pages | 10 |
| Journal | European Journal of Operational Research |
| Volume | 268 |
| Issue number | 2 |
| Early online date | 2 Feb 2018 |
| DOIs | |
| Publication status | Published - 16 Jul 2018 |
Fingerprint
Keywords
- Aggregation
- Bootstrapping
- Forecasting
- Model combination
- Uncertainty
ASJC Scopus subject areas
- Computer Science(all)
- Modelling and Simulation
- Management Science and Operations Research
- Information Systems and Management
Cite this
Exploring the sources of uncertainty: why does bagging for time series forecasting work? / Petropoulos, Fotios; Hyndman, Rob J.; Bergmeir, Christoph.
In: European Journal of Operational Research, Vol. 268, No. 2, 16.07.2018, p. 545-554.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Exploring the sources of uncertainty: why does bagging for time series forecasting work?
AU - Petropoulos, Fotios
AU - Hyndman, Rob J.
AU - Bergmeir, Christoph
PY - 2018/7/16
Y1 - 2018/7/16
N2 - In a recent study, Bergmeir, Hyndman and Benítez (2016) successfully employed a bootstrap aggregation (bagging) technique for improving the performance of exponential smoothing. Each series is Box-Cox transformed, and decomposed by Seasonal and Trend decomposition using Loess (STL); then bootstrapping is applied on the remainder series before the trend and seasonality are added back, and the transformation reversed to create bootstrapped versions of the series. Subsequently, they apply automatic exponential smoothing on the original series and the bootstrapped versions of the series, with the final forecast being the equal-weight combination across all forecasts. In this study we attempt to address the question: why does bagging for time series forecasting work? We assume three sources of uncertainty (model uncertainty, data uncertainty, and parameter uncertainty) and we separately explore the benefits of bagging for time series forecasting for each one of them. Our analysis considers 4004 time series (from the M- and M3-competitions) and two families of models. The results show that the benefits of bagging predominantly originate from the model uncertainty: the fact that different models might be selected as optimal for the bootstrapped series. As such, a suitable weighted combination of the most suitable models should be preferred to selecting a single model.
AB - In a recent study, Bergmeir, Hyndman and Benítez (2016) successfully employed a bootstrap aggregation (bagging) technique for improving the performance of exponential smoothing. Each series is Box-Cox transformed, and decomposed by Seasonal and Trend decomposition using Loess (STL); then bootstrapping is applied on the remainder series before the trend and seasonality are added back, and the transformation reversed to create bootstrapped versions of the series. Subsequently, they apply automatic exponential smoothing on the original series and the bootstrapped versions of the series, with the final forecast being the equal-weight combination across all forecasts. In this study we attempt to address the question: why does bagging for time series forecasting work? We assume three sources of uncertainty (model uncertainty, data uncertainty, and parameter uncertainty) and we separately explore the benefits of bagging for time series forecasting for each one of them. Our analysis considers 4004 time series (from the M- and M3-competitions) and two families of models. The results show that the benefits of bagging predominantly originate from the model uncertainty: the fact that different models might be selected as optimal for the bootstrapped series. As such, a suitable weighted combination of the most suitable models should be preferred to selecting a single model.
KW - Aggregation
KW - Bootstrapping
KW - Forecasting
KW - Model combination
KW - Uncertainty
UR - http://www.scopus.com/inward/record.url?scp=85042179648&partnerID=8YFLogxK
U2 - 10.1016/j.ejor.2018.01.045
DO - 10.1016/j.ejor.2018.01.045
M3 - Article
VL - 268
SP - 545
EP - 554
JO - European Journal of Operational Research
JF - European Journal of Operational Research
SN - 0377-2217
IS - 2
ER -