pith. sign in

arxiv: 2606.04900 · v1 · pith:TSPO65VOnew · submitted 2026-06-03 · 📊 stat.AP

Multi-objective probabilistic forecast combination for inventory demand

Shengjie Wang , Yanfei Kang , Evangelos Spiliotis , Fotios Petropoulos This is my paper

Pith reviewed 2026-06-28 03:52 UTC · model grok-4.3

classification 📊 stat.AP
keywords probabilistic forecast combinationmulti-objective optimizationinventory managementdemand forecastingPareto optimalityforecast accuracycost functions
0
0 comments X

The pith

Multi-objective optimization for probabilistic forecast combinations balances accuracy with inventory decision costs.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper sets out to show that combining probabilistic forecasts can be framed as a multi-objective optimization problem whose solutions trade off statistical accuracy against the actual costs of inventory decisions. Existing single-objective methods optimize only accuracy or only costs, yet the two often conflict under nonlinear inventory penalties. The proposed approach generates a Pareto front of combination weights so that managers can select the point that matches their priorities. Tests on retail sales and spare parts demand data indicate these combinations deliver more balanced results than individual models, simple averages, or single-objective alternatives. If the claim holds, forecasting would be tied directly to the operational metrics that determine profit rather than treated as an isolated statistical task.

Core claim

The paper claims that formulating forecast combination weights as the solution to a multi-objective optimization problem, with one set of objectives drawn from probabilistic scoring rules and the other from inventory cost functions, produces a Pareto-optimal set of combinations that jointly improve accuracy and decision performance relative to single-objective benchmarks.

What carries the argument

Multi-objective optimization that derives the Pareto front of forecast combination weights with respect to both accuracy and inventory cost objectives.

If this is right

  • A set of Pareto-optimal combinations becomes available for explicit selection according to the relative weight a practitioner assigns to accuracy versus cost.
  • Performance remains more balanced across retail sales and spare-parts demand series than is obtained from individual models or single-objective optimization.
  • The framework supplies a direct link between probabilistic forecast quality and the nonlinear holding and shortage costs that govern inventory policy.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same multi-objective structure could be applied to other operational settings where forecast inputs feed into nonlinear costs, such as staffing or pricing.
  • Automated selection rules from the Pareto front could incorporate additional criteria like risk tolerance without changing the underlying optimization.
  • The results underscore the value of constructing cost functions that closely mirror the actual penalty structure faced by the inventory system.

Load-bearing premise

Inventory decision performance can be accurately quantified by explicit cost functions and that selecting from the resulting Pareto front will produce better real-world decisions.

What would settle it

A replication on new demand series in which any selected Pareto-optimal combination produces higher realized inventory costs than the best single-objective or averaging benchmark would falsify the central claim.

Figures

Figures reproduced from arXiv: 2606.04900 by Evangelos Spiliotis, Fotios Petropoulos, Shengjie Wang, Yanfei Kang.

Figure 1
Figure 1. Figure 1: . The process of multi-objective probabilistic forecast combination, including weight [PITH_FULL_IMAGE:figures/full_fig_p017_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: . The relationship between DRPS and cost of M5 data. Each subplot represents a [PITH_FULL_IMAGE:figures/full_fig_p020_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: . Trade-off between holding stock and stockout quantity for the M5 data under three [PITH_FULL_IMAGE:figures/full_fig_p021_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: . The relationship between DRPS and cost for the RAF data. The settings are similar [PITH_FULL_IMAGE:figures/full_fig_p025_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: . Trade-off between holding stock and stockout quantity for the RAF data, with total [PITH_FULL_IMAGE:figures/full_fig_p025_5.png] view at source ↗
read the original abstract

Probabilistic forecasts are essential for inventory management, where decisions depend on the full distribution of future demand. While probabilistic forecast combination is widely used to improve statistical accuracy, most existing approaches optimize statistical loss alone and overlook operational objectives. However, in inventory settings, higher forecast accuracy does not necessarily translate into better decision performance, especially under nonlinear cost structures and multiple, potentially conflicting, decision targets. To address this gap, we propose a multi-objective probabilistic forecast combination framework that simultaneously considers forecast accuracy and inventory decision performance. The framework formulates forecast combination as a multi-objective optimization problem and derives a set of Pareto-optimal combinations, enabling explicit trade-offs between forecasting and operational goals. Empirical studies using Walmart retail data and Royal Air Force spare parts data demonstrate that the proposed approach achieves more balanced and robust performance than individual models, simple averaging, and single-objective optimization. Our results provide a practical and flexible framework for aligning probabilistic forecasting with inventory decision-making.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper proposes a multi-objective probabilistic forecast combination framework for inventory demand. It formulates combination weights as the solution to a multi-objective optimization problem that trades off statistical forecast accuracy against inventory decision costs (holding and shortage), generates the Pareto front of combinations, and selects points from the front. Empirical studies on Walmart retail data and Royal Air Force spare-parts data are reported to show more balanced and robust performance than individual models, simple averaging, and single-objective optimization.

Significance. If the empirical results hold under the stated cost functions, the work is significant for inventory applications because it directly incorporates operational objectives into forecast combination rather than relying on accuracy alone. Credit is due for the use of two real operational datasets and explicit comparison against relevant baselines (individual models, averaging, single-objective). The framework supplies a practical mechanism for explicit trade-offs, which is a clear advance over purely statistical combination methods when decision costs are nonlinear.

major comments (2)
  1. [Empirical studies] Empirical studies section: the central claim that Pareto-optimal combinations yield improved inventory decisions rests on the assumption that the embedded holding/shortage cost functions accurately proxy real operational performance. No out-of-sample validation against actual realized costs or sensitivity checks on cost-parameter misspecification is provided; this directly affects whether the reported balanced performance translates beyond the modeled objective.
  2. [Framework formulation] Framework formulation (optimization problem): the manuscript does not specify how the Pareto front is generated (e.g., exact scalarization method, handling of the probabilistic forecast combination weights, or solver) nor how a practitioner would select a point from the front for deployment. These details are load-bearing for reproducibility and for the robustness claim relative to single-objective baselines.
minor comments (1)
  1. Notation for the multi-objective objectives and the inventory cost parameters should be introduced with explicit definitions and units to improve readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each major comment below and indicate planned revisions to strengthen the manuscript.

read point-by-point responses

  1. Referee: [Empirical studies] Empirical studies section: the central claim that Pareto-optimal combinations yield improved inventory decisions rests on the assumption that the embedded holding/shortage cost functions accurately proxy real operational performance. No out-of-sample validation against actual realized costs or sensitivity checks on cost-parameter misspecification is provided; this directly affects whether the reported balanced performance translates beyond the modeled objective.

    Authors: The paper evaluates combinations on held-out portions of the two real datasets using the modeled holding and shortage costs as the decision metric, which is the conventional approach in this literature when separate realized operational cost records are unavailable. We agree that sensitivity analysis would improve robustness claims and will add experiments that vary the holding-to-shortage cost ratio across a range of values in the revised manuscript. revision: partial

  2. Referee: [Framework formulation] Framework formulation (optimization problem): the manuscript does not specify how the Pareto front is generated (e.g., exact scalarization method, handling of the probabilistic forecast combination weights, or solver) nor how a practitioner would select a point from the front for deployment. These details are load-bearing for reproducibility and for the robustness claim relative to single-objective baselines.

    Authors: We will add a dedicated subsection detailing the scalarization approach (weighted-sum method with a discrete grid of weights), the handling of combination weights on the probabilistic forecasts, the solver employed, and practical selection rules (e.g., knee-point identification or user-specified trade-off). Pseudocode and implementation notes will also be included. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation and claims are self-contained

full rationale

The paper defines a multi-objective optimization framework that trades off statistical forecast accuracy against inventory cost functions, then reports empirical performance on two external datasets (Walmart retail and RAF spare parts). No equation or claim reduces by construction to its own inputs, no fitted parameter is relabeled as a prediction, and no load-bearing premise rests on self-citation. The Pareto front and out-of-sample comparisons are independent of the formulation itself; any performance gain is an empirical observation rather than a definitional identity. This is the normal, non-circular case for an optimization-plus-validation paper.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Based solely on the abstract; the framework depends on the ability to define quantifiable inventory decision objectives and solve the resulting multi-objective problem, but specific parameters and assumptions are not detailed.

free parameters (1)
  • objective trade-off weights
    Weights balancing forecast accuracy against inventory performance objectives must be set or optimized as part of the multi-objective formulation.
axioms (1)
  • domain assumption Inventory costs and decision targets can be accurately modeled as functions of the probabilistic forecasts for use in optimization.
    This premise is required to link the forecast combinations to operational performance.

pith-pipeline@v0.9.1-grok · 5690 in / 1166 out tokens · 54657 ms · 2026-06-28T03:52:15.633109+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

141 extracted references · 65 canonical work pages

  1. [1]

    2004 , doi =

    A new approach to forecasting intermittent demand for service parts inventories , journal =. 2004 , doi =

    2004

  2. [2]

    International Journal of Production Economics , volume=

    Comparison of a new bootstrapping method with parametric approaches for safety stock determination in service parts inventory systems , author=. International Journal of Production Economics , volume=. 2011 , doi=

    2011

  3. [3]

    2012 , doi=

    Snyder, Ralph D and Ord, J Keith and Beaumont, Adrian , journal=. 2012 , doi=

    2012

  4. [4]

    European Journal of Operational Research , volume=

    Combining probabilistic forecasts of intermittent demand , author=. European Journal of Operational Research , volume=. 2024 , doi=

    2024

  5. [5]

    European Journal of Operational Research , volume=

    Huber, Jakob and M. European Journal of Operational Research , volume=. 2019 , doi=

    2019

  6. [6]

    and Eberhart, R

    Kennedy, J. and Eberhart, R. , journal=. Particle swarm optimization , year=

  7. [7]

    Journal of the Operational Research Society , volume=

    An adaptive multi-objective optimal forecast combination and its application for predicting intermittent demand , author=. Journal of the Operational Research Society , volume=. 2024 , doi=

    2024

  8. [8]

    2024 , doi=

    Yang, Wendong and Zang, Xinyi and Wu, Chunying and Hao, Yan , journal=. 2024 , doi=

    2024

  9. [9]

    IEEE Transactions on Evolutionary Computation , volume=

    An evolutionary many-objective optimization algorithm using reference-point-based nondominated sorting approach, part I: solving problems with box constraints , author=. IEEE Transactions on Evolutionary Computation , volume=. 2014 , doi=

    2014

  10. [10]

    IEEE Transactions on Evolutionary Computation , volume=

    An evolutionary many-objective optimization algorithm using reference-point based nondominated sorting approach, part II: Handling constraints and extending to an adaptive approach , author=. IEEE Transactions on Evolutionary Computation , volume=. 2014 , doi=

    2014

  11. [11]

    IEEE Access , title=

    J. IEEE Access , title=. 2020 , volume=

    2020

  12. [12]

    2017 , doi=

    Ganjehkaviri, A and Jaafar, MN Mohd and Hosseini, Seyed Ehsan and Barzegaravval, H , journal=. 2017 , doi=

    2017

  13. [13]

    Management Science , volume =

    Relatively Robust Multicriteria Decisions , author=. Management Science , volume =. 2025 , doi=

    2025

  14. [14]

    2022 , doi=

    Makridakis, S and Spiliotis, E and Assimakopoulos, V , journal=. 2022 , doi=

    2022

  15. [15]

    2024 , doi=

    Olivares-Nadal, Alba V , journal=. 2024 , doi=

    2024

  16. [16]

    Annual Review of Statistics and Its Application , volume=

    Probabilistic forecasting , author=. Annual Review of Statistics and Its Application , volume=. 2014 , publisher=

    2014

  17. [17]

    Combining probabilistic forecasts of

    Taylor, James W and Taylor, Kathryn S , journal=. Combining probabilistic forecasts of. 2023 , publisher=

    2023

  18. [18]

    2024 , doi=

    Van der Meer, Dennis and Pinson, Pierre and Camal, Simon and Kariniotakis, Georges , journal=. 2024 , doi=

    2024

  19. [19]

    Multivariate probabilistic

    Berrisch, Jonathan and Ziel, Florian , journal=. Multivariate probabilistic. 2024 , publisher=

    2024

  20. [20]

    European Journal of Operational Research , year=

    Probabilistic Forecast Aggregation with Statistical Depth , author=. European Journal of Operational Research , year=

  21. [21]

    2023 , doi=

    Wang, Xiaoqian and Hyndman, Rob J and Li, Feng and Kang, Yanfei , journal=. 2023 , doi=

    2023

  22. [22]

    2004 , doi=

    Hora, Stephen C , journal=. 2004 , doi=

    2004

  23. [23]

    Energy , volume=

    A novel combined forecasting model based on neural networks, deep learning approaches, and multi-objective optimization for short-term wind speed forecasting , author=. Energy , volume=. 2022 , doi=

    2022

  24. [24]

    Expert Systems with Applications , volume=

    A novel multivariate combined power load forecasting system based on feature selection and multi-objective intelligent optimization , author=. Expert Systems with Applications , volume=. 2024 , doi=

    2024

  25. [25]

    Computational Economics , volume=

    A novel ensemble learning framework based on news sentiment enhancement and multi-objective optimizer for carbon price forecasting , author=. Computational Economics , volume=. 2025 , doi=

    2025

  26. [26]

    International Journal of Production Economics , volume=

    Optimising forecasting models for inventory planning , author=. International Journal of Production Economics , volume=. 2020 , doi=

    2020

  27. [27]

    2025 , doi =

    European Journal of Operational Research , volume =. 2025 , doi =

    2025

  28. [28]

    2021 , doi =

    International Journal of Production Economics , volume =. 2021 , doi =

    2021

  29. [29]

    European Journal of Operational Research , volume=

    A survey of contextual optimization methods for decision-making under uncertainty , author=. European Journal of Operational Research , volume=. 2025 , doi=

    2025

  30. [30]

    predict, then optimize

    Smart “predict, then optimize” , author=. Management Science , volume=. 2022 , doi=

    2022

  31. [31]

    International Journal of Forecasting , volume=

    Decision-focused linear pooling for probabilistic forecast combination , author=. International Journal of Forecasting , volume=. 2025 , doi=

    2025

  32. [32]

    International Journal of Forecasting , volume=

    Quantile forecast optimal combination to enhance safety stock estimation , author=. International Journal of Forecasting , volume=. 2019 , doi=

    2019

  33. [33]

    International Journal of Forecasting , volume=

    Evaluating predictive count data distributions in retail sales forecasting , author=. International Journal of Forecasting , volume=. 2016 , doi=

    2016

  34. [34]

    International Journal of Forecasting , volume=

    Combining density forecasts , author=. International Journal of Forecasting , volume=. 2007 , doi=

    2007

  35. [35]

    Ensemble forecast of photovoltaic power with online

    Thorey, Jean and Chaussin, Christophe and Mallet, Vivien , journal=. Ensemble forecast of photovoltaic power with online. 2018 , doi=

    2018

  36. [36]

    Journal of Econometrics , volume=

    Dynamic Bayesian predictive synthesis in time series forecasting , author=. Journal of Econometrics , volume=. 2019 , doi=

    2019

  37. [37]

    2022 , doi=

    Goltsos, Thanos E and Syntetos, Aris A and Glock, Christoph H and Ioannou, George , journal=. 2022 , doi=

    2022

  38. [38]

    European Journal of Operational Research , year=

    Fifty Years of Inventory Research from a Forecasting Perspective , author=. European Journal of Operational Research , year=

  39. [39]

    Management Science , volume=

    A practical end-to-end inventory management model with deep learning , author=. Management Science , volume=. 2023 , doi=

    2023

  40. [40]

    Management Science , volume=

    Optimal robust policy for feature-based newsvendor , author=. Management Science , volume=. 2024 , doi=

    2024

  41. [41]

    European Journal of Operational Research , year=

    Integrated estimate-and-optimize decision trees learning for two-stage linear decision-making problems , author=. European Journal of Operational Research , year=

  42. [42]

    Omega , volume=

    A simulation-based multi-objective optimization framework: A case study on inventory management , author=. Omega , volume=. 2017 , doi=

    2017

  43. [43]

    Stone , journal =

    M. Stone , journal =. The Opinion Pool , volume =. 1961 , doi=

    1961

  44. [44]

    Oxford Bulletin of Economics and Statistics , volume=

    Combining density and interval forecasts: a modest proposal , author=. Oxford Bulletin of Economics and Statistics , volume=. 2005 , doi=

    2005

  45. [45]

    2006 , isbn=

    O'Hagan, Anthony and Buck, Caitlin E and Daneshkhah, Alireza and Eiser, J Richard and Garthwaite, Paul H and Jenkinson, David J and Oakley, Jeremy E and Rakow, Tim , title=. 2006 , isbn=

    2006

  46. [46]

    Journal of the Operational Research Society , volume=

    The combination of forecasts , author=. Journal of the Operational Research Society , volume=. 1969 , doi=

    1969

  47. [47]

    Journal of Econometrics , volume=

    Optimal prediction pools , author=. Journal of Econometrics , volume=. 2011 , doi=

    2011

  48. [48]

    International Journal of Forecasting , volume=

    Optimal combination of survey forecasts , author=. International Journal of Forecasting , volume=. 2015 , doi=

    2015

  49. [49]

    Journal of Econometrics , volume=

    Generalised density forecast combinations , author=. Journal of Econometrics , volume=. 2015 , doi=

    2015

  50. [50]

    2016 , doi=

    Del Negro, Marco and Hasegawa, Raiden B and Schorfheide, Frank , journal=. 2016 , doi=

    2016

  51. [51]

    International Journal of Forecasting , volume=

    Bayesian forecast combination using time-varying features , author=. International Journal of Forecasting , volume=. 2023 , doi=

    2023

  52. [52]

    2005 , doi=

    Raftery, Adrian E and Gneiting, Tilmann and Balabdaoui, Fadoua and Polakowski, Michael , journal=. 2005 , doi=

    2005

  53. [53]

    Quarterly Journal of the Royal Meteorological Society , volume=

    Online learning with the continuous ranked probability score for ensemble forecasting , author=. Quarterly Journal of the Royal Meteorological Society , volume=. 2017 , doi=

    2017

  54. [54]

    Journal of Applied Econometrics , volume=

    Combining density forecasts using focused scoring rules , author=. Journal of Applied Econometrics , volume=. 2017 , doi=

    2017

  55. [55]

    2023 , doi=

    Diebold, Francis X and Shin, Minchul and Zhang, Boyuan , journal=. 2023 , doi=

    2023

  56. [56]

    Electronic Journal of Statistics , volume=

    Combining predictive distributions , author=. Electronic Journal of Statistics , volume=. 2013 , doi=

    2013

  57. [57]

    Journal of the Royal Statistical Society: Series B (Statistical Methodology) , volume=

    Combining probability forecasts , author=. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , volume=. 2010 , doi=

    2010

  58. [58]

    International Journal of Forecasting , volume=

    Combining predictive distributions for the statistical post-processing of ensemble forecasts , author=. International Journal of Forecasting , volume=. 2018 , doi =

    2018

  59. [59]

    2003 , doi=

    Garratt, Anthony and Lee, Kevin and Pesaran, M Hashem and Shin, Yongcheol , journal=. 2003 , doi=

    2003

  60. [60]

    Journal of Econometrics , volume=

    Bayesian model averaging and exchange rate forecasts , author=. Journal of Econometrics , volume=. 2008 , doi=

    2008

  61. [61]

    Journal of the American Statistical Association , volume=

    Multivariate Bayesian predictive synthesis in macroeconomic forecasting , author=. Journal of the American Statistical Association , volume=. 2020 , doi=

    2020

  62. [62]

    Journal of the Royal Statistical Society Series B: Statistical Methodology , volume=

    Bayesian predictive decision synthesis , author=. Journal of the Royal Statistical Society Series B: Statistical Methodology , volume=. 2024 , doi=

    2024

  63. [63]

    Management Science , volume=

    Is it better to average probabilities or quantiles? , author=. Management Science , volume=. 2013 , doi=

    2013

  64. [64]

    Management Science , volume=

    Angular combining of forecasts of probability distributions , author=. Management Science , volume=. 2026 , doi=

    2026

  65. [65]

    IISE Transactions , volume=

    Applying deep learning to the newsvendor problem , author=. IISE Transactions , volume=. 2020 , doi=

    2020

  66. [66]

    European Journal of Operational Research , volume=

    Dynamic optimization with side information , author=. European Journal of Operational Research , volume=. 2023 , doi=

    2023

  67. [67]

    2016 , doi=

    Ferreira, Kris Johnson and Lee, Bin Hong Alex and Simchi-Levi, David , journal=. 2016 , doi=

    2016

  68. [68]

    Production and Operations Management , volume=

    Data-driven newsvendor problems regularized by a profit risk constraint , author=. Production and Operations Management , volume=. 2022 , doi=

    2022

  69. [69]

    Advances in Neural Information Processing Systems , volume=

    Task-based end-to-end model learning in stochastic optimization , author=. Advances in Neural Information Processing Systems , volume=

  70. [70]

    Zico , booktitle =

    Donti, Priya and Amos, Brandon and Kolter, J. Zico , booktitle =. Task-based End-to-end Model Learning in Stochastic Optimization , url =

  71. [71]

    Knowledge-Based Systems , volume=

    Constrained multi-objective optimization problems: Methodologies, algorithms and applications , author=. Knowledge-Based Systems , volume=. 2024 , doi=

    2024

  72. [72]

    Journal of the Operational Research Society , volume=

    Forecast combinations for intermittent demand , author=. Journal of the Operational Research Society , volume=. 2015 , doi=

    2015

  73. [73]

    European Journal of Operational Research , volume=

    Elucidate structure in intermittent demand series , author=. European Journal of Operational Research , volume=. 2021 , doi=

    2021

  74. [74]

    Journal of the Operational Research Society , volume=

    Forecasting intermittent demand: a comparative study , author=. Journal of the Operational Research Society , volume=. 2009 , doi=

    2009

  75. [75]

    \ Lerch, S

    baran2018combining APACrefauthors Baran, S. \ Lerch, S. APACrefauthors \ 2018 . Combining predictive distributions for the statistical post-processing of ensemble forecasts Combining predictive distributions for the statistical post-processing of ensemble forecasts . International Journal of Forecasting 34 3 477--496 . APACrefDOI doi:10.1016/j.ijforecast....

    work page doi:10.1016/j.ijforecast.2018.01.005 2018

  76. [76]

    \ Granger, C W

    bates1969combination APACrefauthors Bates, J M. \ Granger, C W. APACrefauthors \ 1969 . The combination of forecasts The combination of forecasts . Journal of the Operational Research Society 20 4 451--468 . APACrefDOI doi:10.1057/jors.1969.103 APACrefDOI

    work page doi:10.1057/jors.1969.103 1969

  77. [77]

    , McCord, C

    bertsimas2023dynamic APACrefauthors Bertsimas, D. , McCord, C. \ Sturt, B. APACrefauthors \ 2023 . Dynamic optimization with side information Dynamic optimization with side information . European Journal of Operational Research 304 2 634--651 . APACrefDOI doi:10.1016/j.ejor.2022.03.030 APACrefDOI

    work page doi:10.1016/j.ejor.2022.03.030 2023

  78. [78]

    \ Deb , K

    pymoo APACrefauthors Blank , J. \ Deb , K. APACrefauthors \ 2020 . Pymoo: Multi-Objective Optimization in Python Pymoo: Multi-Objective Optimization in Python . IEEE Access 8 89497-89509 . APACrefDOI doi:10.1109/ACCESS.2020.2990567 APACrefDOI

    work page doi:10.1109/access.2020.2990567 2020

  79. [79]

    , Jin, M

    chen2025novel APACrefauthors Chen, Y. , Jin, M. , Zhou, Z. \ Tian, Z. APACrefauthors \ 2025 . A novel ensemble learning framework based on news sentiment enhancement and multi-objective optimizer for carbon price forecasting A novel ensemble learning framework based on news sentiment enhancement and multi-objective optimizer for carbon price forecasting ....

    work page doi:10.1007/s10614-024-10828-6 2025

  80. [80]

    , De Mol, C

    conflitti2015optimal APACrefauthors Conflitti, C. , De Mol, C. \ Giannone, D. APACrefauthors \ 2015 . Optimal combination of survey forecasts Optimal combination of survey forecasts . International Journal of Forecasting 31 4 1096--1103 . APACrefDOI doi:10.1016/j.ijforecast.2015.03.009 APACrefDOI

    work page doi:10.1016/j.ijforecast.2015.03.009 2015

Showing first 80 references.