We use classification methods from machine learning to predict the destination of global crude oil exports by utilising micro-level crude oil shipment data that incorporates attributes related to the contract, cargo specifications, vessel specifications and macroeconomic conditions. The results show that micro-level information about the oil shipment such as quality and cargo size dominates in the destination prediction. We contribute to the academic literature by providing the first machine learning application to oil shipment data, and by providing new knowledge on the determinants of global crude oil flows. The machine-learning models used to predict the importing country can reach an accuracy of above 71% for the major oil exporting countries based on out-of-sample tests and outperform both naive models and discrete regression models.
AdlandR.JiaH.StrandenesS.P. (2017). “Are AIS-based trade volume estimates reliable? The case of crude oil exports.” Maritime Policy and Management44(5): 657-665. https://doi.org/10.1080/03088839.2017.1309470.
2.
AllenbyG.M.RossiP.E. (1998). “Marketing models of consumer heterogeneity.” Journal of Econometrics 89 (1): 5778. https://doi.org/10.1016/S0304-4076(98)00055-4.
3.
AlizadehA.H.StrandenesS.PThanopoulouH. (2016). “Capacity retirement in the dry bulk market: a vessel based logit model.” Transportation Research Part E92: 28-42. https://doi.org/10.1016/j.tre.2016.03.005.
4.
AndrewsR.L.AinslieA.CurrimI.S. (2002). “An empirical comparison of logit choice models with discrete versus continuous representations of heterogeneity.” Journal of Marketing Research39(4): 479-487. https://doi.org/10.1509/jmkr.39.4.479.19124.
5.
BellmanR.E. (1961). Adaptive control processes: a guided tour. Princeton University Press. https://doi.org/10.1515/9781400874668.
6.
BishopC.M. (2006). Pattern Recognition and Machine Learning. Springer, ISBN 978-0-387-31073-2.
7.
BreimanL. (1984). Classification and regression trees, New York: Routledge.
8.
BreimanL.FriedmanJ.H.OlshenR.A.StoneC.J. (1998). Classification and regression trees. Chapman & Hall/CRC. Breiman, L. (2001). “Random Forests.” Machine Learning45: 5-32. https://doi.org/10.1023/A:1010933404324.
9.
ChenT.GuestrinC.. (2016). “Xgboost: A scalable tree boosting system.” In Proceedings of the 22ndacm sigkdd international conference on knowledge discovery and data mining (pp. 785-794). ACM. https://doi.org/10.1145/2939672.2939785.
10.
CIA (2018). Central Intelligence Agency, the World Factbook. https://www.cia.gov/library/publications/resources/the-world-factbook/fields/261.html.
11.
CirilloC.XuR. (2011). “Dynamic discrete choice models for transportation.” Transport Reviews: 31(4): 473-494.https://doi.org/10.1080/01441647.2010.533393.
12.
Clarkson Research (2016). Shipping Intelligence Network, www.clarksons.net.
13.
DoyleJ., (1994). Crude Awakening: The Oil Mess in America: Wasting Energy Jobs and the Environment. Friends of the Earth, Washington, DC.
14.
CruzJ.A.WishartD.W. (2006). “Applications of machine learning in cancer prediction and prognosis.” Cancer Informatics2: 59-78. https://doi.org/10.1177/117693510600200030.
15.
FreitagD. (2000). “Machine Learning for Information Extraction in Informal Domains.”Machine Learning 39: 169-202. https://doi.org/10.1023/A:1007601113994.
16.
FriedmanJ.H. (1999a). Greedy Function Approximation: A Gradient Boosting Machine. IMS 1999 Reitz lecture https://statweb.stanford.edu/∼jhf/ftp/trebst.pdf (accessed 1/8/2019).
FriedmanJ.H. (2001). “Greedy function approximation: a gradient boosting machine.” Annals of statistics 1189-1232. https://doi.org/10.1214/aos/1013203451.
GreeneW.H.HensherD.A. (2003). “A latent class model for discrete choice analysis: Contrasts with mixed logit.” Transportation Research Part B37(8): 681-698. https://doi.org/10.1016/S0191-2615(02)00046-2.
21.
GroverP. (2017). Gradient boosting from scratch. https://medium.com/mlreview/gradient-boosting-from-scratch-1e317ae4587d.
22.
HensherD.GreeneW. (2003). “The mixe3d logit model: the state of practice.” Transportation30(2): 133-176. https://doi.org/10.1023/A:1022558715350.
23.
IEA (2019). International Energy Agency, Key world energy statistics. https://www.iea.org/statistics/kwes/.
24.
JiaH. (2018). “Crude oil trade and green shipping choices.” Transportation Research Part D65: 618-634. https://doi.org/10.1016/j.trd.2018.10.003.
25.
JiaH.AdlandR.PrakashV.SmithT. (2017). “Energy efficiency with the application of virtual arrival policy.” Transportation Research Part D54: 50-60. https://doi.org/10.1016/j.trd.2017.04.037.
26.
KampichlerC.WielandR.CalmeS.WeissenbergerH.Arriaga-WeissS. (2010). “Classification in conservation biology: A comparison of five machine-learning methods.” Ecological Informatics5(6): 441-50. https://doi.org/10.1016/j.ecoinf.2010.06.003.
27.
KearnsM. (1988). Thoughts on hypothesis boosting. https://www.cis.upenn.edu/∼mkearns/papers/boostnote.pdf.
28.
KotsiantisS.B. (2007). Supervised machine learning: a review of classification techniques.
29.
KubatM.HolteR.C.MatwinS. (1998). “Machine learning for the detection of oil spills in satellite radar images.” Machine Learning30: 195-215. https://doi.org/10.1023/A:1007452223027.
30.
ManskiC.F.McFaddenD.L. (1981). A structural analysis of discrete data with econometric applications. Cambridge: the MIT Press.
31.
MalchowM.B.KanafaniA. (2004). “A disaggregate analysis of port selection.” Transportation Research Part E40: 317-337. https://doi.org/10.1016/j.tre.2003.05.001.
32.
McFaddenD. (1974). “Conditional logit analysis of qualitative choice behavior.” In P. Zarembhk (ed.). Frontiers in Econometrics. Academic Press: Berkeley, California.
33.
McFaddenD. (1987). “Regression-based specification tests for the multinomial logit model.” Journal of Econometrics34: 63-82. https://doi.org/10.1016/0304-4076(87)90067-4.
34.
McFaddenD. (1989). “A method of simulated moments for estimation of discrete response models without numerical integration.” Econometrica57(5): 995-1026. https://doi.org/10.2307/1913621.
35.
McFaddenD.TrainK. (2000). “Mixed MNL models for discrete response.” Journal of Applied Econometrics15: 447470. https://doi.org/10.1002/1099-1255(200009/10)15:5<447::AID-JAE570>3.0.CO;2-1.
36.
McLachlanG.PeelD. (2005). Finite mixture models. John Wiley & Sons.
37.
OPEC (2016). OPEC Share of World Crude Oil Reserve, http://www.opec.org/opec_web/en/data_graphs/330.htm.
38.
PedregosaF.VaroquauxG.GramfortA.MichelV.ThirionB.GriselO.VanderplasJ. (2011). “Scikit-learn: Machine learning in Python.” Journal of machine learning research 12(Oct), 2825-2830.
39.
PiendlR.LiedtkeG.MatteisT. (2017). “A logit model for shipment size choice with latent classes—empirical findings for Germany.” Transportation Research Part A102: 188-201. https://doi.org/10.1016/j.tra.2016.08.023.
40.
RichJ.HolmbladP.M.HansenC.O. (2009). “A weighted logit freight mode-choice model.” Transportation research Part E45: 1006-1019. https://doi.org/10.1016/j.tre.2009.02.001.
41.
SamuelA.L. (1959). “Some studies in machine learning using the game of checkers.” IBM Journal3(3): 535-554. https://doi.org/10.1147/rd.33.0210.
42.
SteveA.B.CorsiT.M. (2012). “Choosing a port: an analysis of containerized imports into the U.S.” Transportation Research Part E48: 881-895. https://doi.org/10.1016/j.tre.2012.02.003.
43.
SumermansH.J. (1996). “A stated choice model of sequential mode and destination choice behaviour for shopping trips.” Environment and Planning A28(1): 173-184. https://doi.org/10.1068/a280173.
44.
TrainK. (1986). Qualitative Choice Analysis: Theory, Econometrics, and an Application to Automobile Demand. MIT Press. Chapter 8.
45.
TrainK. (2008). “EM Algorithms for nonparametric estimation of mixing distributions.” Journal of Choice Model1(1): 40-69. https://doi.org/10.1016/S1755-5345(13)70022-8.
46.
TrainK.WinstonC. (2007). “Vehicle Choice Behavior and the Declining Market Share of U.S. Automakers.” International Economic Review48(4): 1469-1496. https://doi.org/10.1111/j.1468-2354.2007.00471.x.
47.
VapnikV. (1995). The Nature of Statistical Learning Theory (information science and statistics). Springer Verlag. https://doi.org/10.1007/978-1-4757-2440-0.
48.
VijA.KruegerR. (2017). “Random taste heterogeneity in discrete choice models: flexible nonparametric finite mixture distributions.” Transportation Research PartB.106: 76-101. https://doi.org/10.1016/j.trb.2017.10.013.
49.
WangL.and LoL. (2007). “Immigrant grocery-shopping behavior: Ethnic identity versus accessibility.” Environment and Planning A39(3): 684-699. https://doi.org/10.1068/a3833.
50.
WittenI.H.FrankE. (2005) Data Mining Practical Machine Learning Tools and Techniques. San Francisco: Elsevier Inc.
51.
WuX.KumanV.QuinlanJ.R.GhoshJ.YangQ.MotodaH.McLachlanG.J.NgA.LiuB.YuPS.ZhouZ.Stein-bachM.HandD.J.SteinbergD. (2008). “Top 10 algorithms in data mining.” Knowledge and Information System.14(1): 1-37. https://doi.org/10.1007/s10115-007-0114-2.
