Paper: arXiv 2305.16255
Authors: Paul Ghelasi, Florian Ziel
Abstract
Aggregated curves are common structures in economics and finance, and the most prominent examples are supply and demand curves. In this study, we exploit the fact that all aggregated curves have an intrinsic hierarchical structure, and thus hierarchical reconciliation methods can be used to improve the forecast accuracy. We provide an in-depth theory on how aggregated curves can be constructed or deconstructed, and conclude that these methods are equivalent under weak assumptions. We consider multiple reconciliation methods for aggregated curves, including previously established bottom-up, top-down, and linear optimal reconciliation approaches. We also present a new benchmark reconciliation method called ‘aggregated-down’ with similar complexity to bottom-up and top-down approaches, but it tends to provide better accuracy in this setup. We conducted an empirical forecasting study on the German day-ahead power auction market by predicting the demand and supply curves, where their equilibrium determines the electricity price for the next day. Our results demonstrate that hierarchical reconciliation methods can be used to improve the forecasting accuracy of aggregated curves.
Complexity vs Empirical Score
- Math Complexity: 8.0/10
- Empirical Rigor: 6.5/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper introduces significant mathematical complexity through the formalization of aggregated curves as hierarchical structures, deriving matrix representations and equivalence results, while also presenting a dedicated empirical forecasting study on the German electricity market with specific accuracy metrics and implementation details.
Research Flowchart
flowchart TD A["Research Goal:<br>Improve forecast accuracy for<br>aggregated curves via hierarchical methods"] --> B["Data Source:<br>German Day-Ahead Power Auction<br>Demand & Supply Curves"] B --> C["Methodology:<br>Apply Hierarchical Reconciliation<br>Bottom-Up, Top-Down, Linear Optimal, Aggregated-Down"] C --> D["Computational Process:<br>Reconstruct equilibrium prices<br>from reconciled forecasts"] D --> E["Key Findings:<br>1. Hierarchical methods improve accuracy<br>2. New 'Aggregated-Down' benchmark<br>3. Theory proves equivalence of construction/deconstruction"]
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper introduces hierarchical reconciliation methods for aggregated curves and a new benchmark method called aggregated-down to improve forecast accuracy.
Source sentence
We also present a new benchmark reconciliation method called 'aggregated-down' with similar complexity to bottom-up and top-down approaches, but it tends to provide better accuracy in this setup.
Assumes 3
- Equivalence of construction and deconstruction methods holds under weak assumptions. model
Source sentence
We provide an in-depth theory on how aggregated curves can be constructed or deconstructed, and conclude that these methods are equivalent under weak assumptions.
- Proportions calculated using values that are closer in the hierarchy are more accurate. model
Source sentence
The motivation for this approach is based on the simple assumption that proportions calculated using values that are closer in the hierarchical structure are expected to be more accurate than when using values that are further away.
- Forecasting algorithm is invariant to orthogonal transformations (e.g., linear regressions). model
Source sentence
the assumption by = B[k]by[k] is satisfied if the forecasting algorithm that provides by is invariant to orthogonal transformations, e.g., this holds for linear regressions.
Data
- German day-ahead electricity market supply and demand curves market-prices public
Source sentence
We conducted an empirical forecasting study on the German day-ahead power auction market by predicting the demand and supply curves, where their equilibrium determines the electricity price for the next day.
- Price grid data for 5 May 2020 market-prices public
Source sentence
corresponding 24 supply and demand curves on the price grid between -60 and 120 EUR/MWh for 5 May 2020 with highlighted curves for 11:00
Costs
Not applicable
Finds 3
- Hierarchical reconciliation methods improve forecasting accuracy of aggregated curves. predictive-accuracy
Source sentence
Our results demonstrate that hierarchical reconciliation methods can be used to improve the forecasting accuracy of aggregated curves.
- Optimal weighted least squares (WLS) reconciliation yields the lowest errors for the largest number of price classes. performance
Source sentence
The optimal weighted least squares (WLS) reconciliation approach yielded the lowest errors for the largest number of classes, followed by the shrinkage and lambda approaches.
- The aggregated-down approach, a localized top-down method, performs well in simulations supporting its design assumption. theoretical
Source sentence
The new aggregated-down approach is motivated based on the simple idea of closeness, i.e. proportions calculated using values that are closer in the hierarchical structure are expected to be more accurate than those calcula…
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 9 related papers). Coverage is partial, so this is neither replication nor refutation.
2 papers in the archive
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.