Intelligent Decision Making for Commodities Price Prediction: Opportunities, Challenges and Future Avenues
The DOI matches its Crossref record; 127 of the 127 references it cites were found in a registry or an index; 11 of the 60 sources its claims cite were checked for support.
Overview
CoChat Summary
TL;DR
This systematic review surveys machine-learning and deep-learning models for commodity price prediction, focusing on house, stock, natural gas, and crude oil prices. It finds that ML/DL methods show promise, especially for house price prediction, but performance varies and is constrained by data quality, feature selection, overfitting, interpretability, and limited cross-market integration.
Key findings
- The review screened 150 studies, removed 50 duplicates, refined 120, and included 107 eligible articles (2008–2024) covering house, stock, natural gas, and crude oil price prediction.
- Reported results vary widely across included works: an ANN for Singapore condominium prices achieved RMSE 0.732 and R-value 0.983; XGBoost for Chengdu second-hand house prices reached 0.9251 accuracy; gradient boosting for Vijayawada property prices reached 91% accuracy; and LSTM for NIFTY 50 stock prediction reached 83.88% accuracy.
- For natural gas, a DNN using 281 months achieved MSE 0.0595, RMSE 0.2440, and R² 0.9937; an ARNN reduced MSE by 33% versus ARIMA. For crude oil, a 1-D CNN achieved F1 97.90% and AUC 0.99 for contamination detection, and an L-GCN model outperformed traditional methods for traffic flow.
- The review concludes that machine learning holds significant potential for optimizing house price predictions, but existing studies often analyze house, stock, and natural gas prices independently, ignoring dynamic interdependencies; hybrid/integrated models such as VAR combined with ML are recommended.
Methods
- Systematic literature review guided by five research questions on ML techniques, key features, preprocessing, challenges, and evaluation metrics.
- Searched Scopus, Elsevier, ACM, IEEE Xplore, and Google Scholar; inclusion criteria covered 2008–2024, English-language, full-text, peer-reviewed conference or journal papers using ML, deep learning, or fuzzy logic in commodity prediction; excluded non-peer-reviewed work, theses, and non-ML papers.
- From 150 identified studies, 50 duplicates were removed, 120 were refined, and 107 eligible articles were included; findings were synthesized narratively and in comparative tables by commodity category.
Limitations
- Stated by the authors: limited data availability; data quality and reliability issues such as missing values, outliers, or biased samples; lack of interpretability; overfitting and model complexity; limited generalizability across regions; lack of causal inference; changing market dynamics; and data privacy/security concerns.
- Evident from the text: the review is secondary and reports no primary experiments; many included studies have small or narrow datasets, insufficient feature discussion, missing comparative analysis, and limited external validation, and the review itself notes that the literature often lacks rigorous feature selection, preprocessing, and overfitting mitigation.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
10 Valid · 3 Not valid · 42 Not checked · 32 Own results and interpretations
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“The studies Peng et al. (2021), Ma et al. (2021), Oust et al. (2020) propose LUCE, the first automated property valuation model that is life-long predictive.”
What the sources say
- Supportedcites Peng et al. (2021) matched to “Lifelong Property Price Prediction: A Case Study for the Toronto Real Estate Market” (2021) source
Peng et al. present LUCE as their model in the opening passage (p. 1), explicitly calling it “the first life-long predictive model for automated property valuation.” The conclusion (p. 13) also describes LUCE as a lifelong learning framework for house valuation. This verdict concerns the attribution to Peng et al., not the separately named studies. (LLM, conf=0.94)
- Not matchedcites Ma et al. (2021)
This doesn’t read as a citation we can look up.
- Not checked — no full text of the sourcecites Oust et al. (2020) matched to “Combining Property Price Predictions from Repeat Sales and Spatially Enhanced Hedonic Regressions” (2020) source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: unclear — It is unclear why the paper says it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“the analysis shows that the suggested method outper- forms the current models by a wide margin.”
What the sources say
- Partly supportedcites Wang et al., 2021 matched to “Deep Learning Model for House Price Prediction Using Heterogeneous Data Analysis Along With Joint Self-Attention Mechanism” (2021) sourceExperimental — accuracy not yet measured
Section D reports that the paper’s proposed joint self-attention model had better average performance than the benchmarks, and Section F says it outperformed the other models tested. The paper identifies its method as Joint Self-Attention, not JGC_MMN, and these passages do not establish that it won by a wide margin. Page numbers were not available in our copy. (LLM, conf=0.82)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“The evaluation con- ducted on the UCI Machine Learning Repository Boston housing dataset, consisting of 14 features and 506 entries, demonstrates that the proposed model achieves an acceptable level of accuracy with a prediction error margin of ± 5 when compared to actual prices.”
What the sources say
- Supportedcites Adetunji et al., 2022 matched to “House Price Prediction using Random Forest Machine Learning Technique” (2022) source
The Abstract states that the model was evaluated on the UCI Boston housing dataset with 506 entries and 14 features, and that its predictions had an acceptable error margin of ±5. Section 3.2 likewise reports that differences between predicted and actual prices were within ±5. (LLM, conf=0.95)
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
In the paper
“As a result, employing several stock indices, LSTM was implemented and compared to SVR, demonstrating LSTM's greater accuracy.”
What the sources say
- Not checked — no full text of the sourcecites Bathla, 2020 matched to “Stock Price prediction using LSTM and SVR” (2020) source
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on an assumption not shown here.
In the paper
“While some market assumptions believe that exact stock price forecasting is impossible, (Sisodia et al., 2022) shows that it may be done to some extent.”
What the sources say
- Not matchedcites Sisodia et al., 2022
An author–year citation the paper’s reference list doesn’t settle. We don’t search these by name.
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
26 more claims
Create a free account to see all 87 claims and to verify the 5 references still open.
- Probably not a substantial claim. 45
- It may state two separate claims as one. 21
- The quote only partly states this claim. 2
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
“A total of 150 studies are identified through library searches including Google Scholar (n = 30), IEEE Xplore (n = 50), Elsevier (n=10), ACM (n=50), and Scopus (n = 10). After removing duplicates (n =50), we refine the remaining 120 research works, and 107 eligible related literature articles are included in this review”
Reported 120 · recomputed 100
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Abstract
“Published in academic conferences or peer-reviewed journals.”
§ Abstract
“Papers that are not published via peer-reviewed journals.”
Citations
Cites
Journal of Physics: Conference Series
(2013) · Journal of Physics Conference Series · doi:10.1088/issn.1742-6596 · Cited by 15,679
A Quick Review of Machine Learning Algorithms
Ray, S. (2019) · 2019 International Conference on Machine Learning, Big Data, Cloud and Parallel Computing (COMITCon) · doi:10.1109/comitcon.2019.8862451 · Cited by 1,070
IEEE Transactions on Instrumentation and Measurement
(2023) · doi:10.1109/tim.19 · Cited by 503
Park, B., & Bae, J. K. (2015) · Expert Systems with Applications · doi:10.1016/j.eswa.2014.11.040 · Cited by 401
Determinants of house prices in Turkey: Hedonic regression versus artificial neural network
Selim, H. (2009) · Expert Systems with Applications · doi:10.1016/j.eswa.2008.01.044 · Cited by 362
IEEE Transactions on Knowledge and Data Engineering
(2004) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2004.1269660 · Cited by 301
Hu, L., He, S., Han, Z., Xiao, H., Su, S., Weng, M., & Cai, Z. (2019) · Land Use Policy · doi:10.1016/j.landusepol.2018.12.030 · Cited by 259
Čeh, M., Kilibarda, M., Lisec, A., & Bajat, B. (2018) · ISPRS International Journal of Geo-Information · doi:10.3390/ijgi7050168 · Cited by 206
Understanding house price appreciation using multi-source big geo-data and machine learning
Kang, Y., Zhang, F., Peng, W., Gao, S., Rao, J., Duarte, F., & Ratti, C. (2021) · Land Use Policy · doi:10.1016/j.landusepol.2020.104919 · Cited by 205
Take a Look Around: Using Street View and Satellite Images to Estimate House Prices
Law, S., Paige, B., & Russell, C. (2019) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3342240 · Cited by 203
Research works printed but not linked: 1
- Géron, A. (2022). Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow. O'Reilly Media Inc.
Cited by
Quantum Finance: Exploring the Implications of Quantum Computing on Financial Models
Zhou, J. (2025) · Computational Economics · doi:10.1007/s10614-025-10894-4 · Cited by 78
Nazir, I., Mushtaq, N., & Amin, W. (2025) · Energies · doi:10.3390/en18195076 · Cited by 18
Multi-Source Stock Price Prediction Using BiLSTM-GRU-RF with Sentiment and Macroeconomic Indicators
Pushpavathi, V., Bhaskar, N., Maheswari, K., Madhukar, G., Madhusudhan, M., & Lavanya, G. (2025) · IEEE International Conference on Intelligent Signal Processing and Effective Communication Technologies (INSPECT) · doi:10.1109/inspect67393.2025.11350457 · Cited by 1
Govardhanan, K. (2025) · International Conference on Applied Artificial Intelligence and Computing (ICAAIC) · doi:10.1109/icaaic64647.2025.11331171
Metadata
- Authors
- Saeed, N., Shafi, I., Pervez, S., Thompson, E. B., Castilla, A. K., Samad, M. A., & Ashraf, I.
- Affiliations
- National University of Sciences and Technology (PK)
- Iqra University (PK)
- Universidad Europea del Atlántico (ES)
- Yeungnam University (KR)
- Corresponding author
- Samad, Md Abdus
- Ashraf, Imran
- Venue
- Computational Economics
- Published
- 2025-01-07
- Volume
- 66
- Issue
- 5
- Pages
- 3781-3839
- ISSN
- 0927-7099 (ISSN-L)
- 1572-9974
- OpenAlex
- W4406149304
- Semantic Scholar
- 014b3ca65429a9206e40f9b5751654e65bb257cd
- Cited by
- 11
- References
- 122
- Citations by year
- 2026: 5 · 2025: 6
- FWCI
- 5.54
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- www.springernature.com/gp/researchers/text-and-data-mining (tdm)
- www.springernature.com/gp/researchers/text-and-data-mining (vor)
- Locations
- Computational Economics (publishedVersion)
- Funders
- the European University of Atlantic
- Topics
- Social Sciences › Decision Sciences › Management Science and Operations Research › Stock Market Forecasting Methods
- Social Sciences › Decision Sciences › Management Science and Operations Research › Forecasting Techniques and Applications
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Energy Load and Power Forecasting
- Keywords
- commodity price forecasting, house price prediction, stock price prediction, ensemble methods, artificial neural networks, support vector machine, time series analysis, regression analysis, machine learning, risk management, real estate
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Intelligent Decision Making for Commodities Price Prediction: Opportunities, Challenges and Future Avenues. https://app.cochat.ai/papers/10.1007/s10614-024-10837-5
- The paper
- Saeed, N., Shafi, I., Pervez, S., Thompson, E. B., Castilla, A. K., Samad, M. A., & Ashraf, I. (2025). Intelligent Decision Making for Commodities Price Prediction: Opportunities, Challenges and Future Avenues. Computational Economics. https://doi.org/10.1007/s10614-024-10837-5
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.