Information aggregation based trend prediction of energy structure via an improved compositional data time series forecasting model and its application
Jingjing Ma · 2026
Read Information aggregation based trend prediction of energy structure via an improved compositional data time series forecasting model and its application by Jingjing Ma (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0351310.txt.
Description
Jingjing Ma refines how forecasters combine predictions about a country's energy mix. The 2026 study works with compositional data: shares of coal, oil, gas, and renewables that must sum to a whole. Single forecasting models shine at different times. Combining them helps, but only if the weights adapt. Ma proposes an aggregation method built on an ordered weighting operator, informed by each model's recent accuracy. A genetic algorithm searches for the weight set that minimises error against the real series.
Two combination schemes resulted: one with time-varying weights, one built on the new operator. Both were tested on China's primary energy production structure. The varying-weight model beat the benchmark on prediction accuracy. Energy planners live and die by share forecasts: grid investment, import contracts, and carbon budgets all hang on them. A demonstrably better way to blend rival forecasts is a quiet but consequential upgrade.
Catalog record
| CREATOR | Jingjing Ma |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 40216 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |