A novel approach for panel data: An ensemble of weighted functional margin SVM models

Erdogan, Bi̇rsen Eygi; Özöğür‐Akyüz, Süreyya; Ataş, Pınar Karadayı

doi:10.1016/j.ins.2019.02.045

Cited by 21 publications

(7 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A multi-objective optimization problem could be developed, taking into consideration the time-varying stochastic price-dependent demand (Ghoreishi et al, (2015)) and robust algorithms can be developed as solution methodologies (Kara et al, (2019), Pervin et al, (2018, Roy et al, (2017), andÖzöugür et al, (2016)). Moreover, machine learning algorithms (Alkahtani, et al, (2019) including regression methodologies ( (Taylan et al, 2019b) and (Taylan et al, 2019a)), artificial neural networks and support vector machines ( (Erdogan et al, 2019), (Savku et al, (2014) and Savku et al, (2018)), modern optimization techniques (Alp et al, (2011)) and environmentally conscious optimization (Paksoy et al, 2014) can be used for predicting the fuel pricing scenarios and bunker fuel prices at different port of calls based on the historical data. In future, this research could also be extended by using weather data to estimate the real-time fuel consumption which may differ depending upon the weather conditions in the voyage.…”

Section: Discussionmentioning

confidence: 99%

Bunkering policies for a fuel bunker management problem for liner shipping networks

Choudhary

Türkay

et al. 2021

European Journal of Operational Research

View full text Add to dashboard Cite

This paper investigates the problem of bunker fuel management for liner shipping networks under different fuel pricing scenarios and taking into consideration different fuel bunkering policies. The fuel consumption of a vessel on a sailing leg may fluctuate as the real vessel speed deviates from the planned vessel speed. Furthermore, fluctuation of fuel prices at various ports increases the complexity of bunkering decisions related to the selection of the bunkering ports and the estimation of bunkered fuel cost.We have developed a mixed integer non-linear programming model to minimise the total expected cost consisting of inventory cost related to container transportation, operating cost associated with ship hiring, as well as bunkering cost and fuel consumption cost at the port. The novelty of our research lies in its consideration of stochastic fuel consumption for different sailing legs, stochastic fuel prices at each port and different fuel bunkering policies to determine optimal bunker fuel management strategies for the selection of bunkering ports and for the estimation of the amount of bunkered fuel required.We have proposed a novel approximate algorithm based on mathematical formulation and the fuel bunkering policies to calculate the total expected cost; the fuel inventory while arriving at and departing from the port; the number of vessels hired for weekly service; the arrival and departure time of the ship; and the amount of fuel bunkered at a port. We have performed extensive computational experiments on the practical routes to demonstrate the applicability, efficacy and robustness of the proposed novel methodology.

show abstract

Section: Discussionmentioning

confidence: 99%

Bunkering policies for a fuel bunker management problem for liner shipping networks

Choudhary

Türkay

et al. 2021

European Journal of Operational Research

View full text Add to dashboard Cite

show abstract

“…The typical machine learning models for default prediction mainly include SVM, KNN, DT, and RF. Erdogan et al (2021) and Perols et al (2017) proposed predicting credit rating method based on SVM. Shen et al (2020) constructed a dynamic financial distress forecast model with multiple forecast results under unbalanced data environment based on RF.…”

Section: Literature Reviewmentioning

confidence: 99%

EWT‐SMOTE to improve default prediction performance in imbalanced data: Analysis of Chinese data

Zhou,

Lin,

Chi

et al. 2023

Journal of Forecasting

View full text Add to dashboard Cite

This study aims to solve the imbalanced sample problem in default prediction. We calculate the classification contribution score of each default customer by the entropy weight technique (EWT) for order of preference by similarity to the ideal solution and construct a default prediction model according to several models. Our proposed EWT‐synthetic minority oversampling technique (SMOTE) method significantly improves the prediction accuracy of several typical default prediction models and reduces type II error. We find that the indicators “net cash flow from operating activities,” “Engel coefficient,” “basic earnings per share,” and “total social retail sales” significantly influence default prediction of Chinese listed companies.

show abstract

“…In general, traditional enterprise risk analysis methods mainly consider financial indicators, such as profitability index, operating efficiency and solvency, using multivariate discriminant analysis [2,24,28], or machine learning methods, such as SVM and Decision Tree [6,9,23,34]. Recently, much research has concentrated on utilizing text information, such as financial reports and conference calls, for mining enterprise intra-risk [5,15].…”

Section: Related Workmentioning

confidence: 99%

Combining Intra-Risk and Contagion Risk for Enterprise Bankruptcy Prediction Using Graph Neural Networks

Yu¹,

Sang²,

Yu³

et al. 2022

Preprint

View full text Add to dashboard Cite

Bankruptcy risk prediction for Small and Medium-sized Enterprises (SMEs) is a crucial step for financial institutions to make the loan decision and identify region economics's early warning. However, previous studies in both finance and AI research fields only consider either the intra-risk or the spillover-risk, ignoring their interactions and their combinatorial effect for simplicity. This paper for the first time considers both risks simultaneously and their joint effect in bankruptcy prediction. Specifically, we first propose an enterprise intra-risk encoder with LSTM based on enterprise risk statistical significance indicators from its basic business information and litigation information for its intra-risk learning. Afterward, we propose an enterprise spillover-risk encoder based on enterprise relational information from the enterprise knowledge graph for its spillover-risk embedding. In particular, the spillover-risk encoder is equipped with both the newly proposed Hyper-Graph Neural Networks (Hyper-GNNs) and Heterogeneous Graph Neural Networks (Heter-GNNs), which is able to model spillover risk from two different aspects, i.e. common risk factors based on hyperedges and direct diffusion risk from the neighbors, respectively. With the two kinds of encoders, a unified framework is designed to simultaneously capture intra-risk and spillover-risk for bankruptcy prediction. To evaluate our model, we collect multi-sources SMEs real-world data and build a novel benchmark dataset SMEsD. We provide open access to the dataset, which is expected to promote the financial risk analysis research further. Experiments on SMEsD against nine SOTA baselines demonstrate the effectiveness of the proposed model for bankruptcy prediction.

show abstract

A novel approach for panel data: An ensemble of weighted functional margin SVM models

Cited by 21 publications

References 24 publications

Bunkering policies for a fuel bunker management problem for liner shipping networks

Bunkering policies for a fuel bunker management problem for liner shipping networks

EWT‐SMOTE to improve default prediction performance in imbalanced data: Analysis of Chinese data

Combining Intra-Risk and Contagion Risk for Enterprise Bankruptcy Prediction Using Graph Neural Networks

Contact Info

Product

Resources

About