본문 바로가기
KERD
논문KCI 등재

ARIMA-SVM 하이브리드 모델을 활용한 탱커 운임지수 예측

Forecasting Tanker Indices with ARIMA-SVM Hybrid Models

0

조회

0

다운로드

3

인용

0

참고문헌

키워드

주제분류경영학

초록

탱커시장은 스팟거래의 비중이 상당히 높은 시장으로서 건화물시장이나 컨테이너시장과 달리 단기예측의 중요성이 강조된다. 본 논문은 탱커시황 예측에 기계학습 모델(ANN과 SVM)과 계량모델(ARIMA)을 결합한 하이브리드 모델(Zhang, 2003)을 적용하였다. 탱커시장을 대표하는 BDTI와 BCTI를 연구대상으로 하였으며 연구자료는 1998년 8월부터 2018년 7월까지 240개월이다. 변동성을 고려하여 연구기간을 하위구간 1(1998년∼2008년), 하위구간 2(2009년∼2018년), 전체구간으로 구분하였고 모형의 예측성과를 비교하였다. 단기비중이 높은 거래의 형태를 감안하여 월별예측을 시도하였는데 이는 스팟거래가 20∼50일을 기준으로 하는 계약이기 때문이다. 탱커시황 예측결과는 본 논문에서 제시한 ARIMA-SVM의 모델이 다른 모델들에 비해 성능이 우수하였다. 본 연구는 BDTI와 BCTI를 대상으로 하여 전체 탱커시장을 포괄할 뿐만 아니라 탱커시장에서 비중이 높은 스팟거래의 기간과도 부합하는 결과를 제시하였기 때문에 실무적인 성과로서 의의가 있다. 또한 학문적으로 해운시황예측에 시도되지 않았던 하이브리드 모델이라는 방법론을 제시하였다는 점에서 의미를 찾을 수 있다. 본 논문의 결과는 스팟거래 비중이 높은 탱커시장에 단기예측에 관한 과학적 근거로 제시되어 시장참여자들의 용선의사결정에 실질적으로 도움을 줄 것으로 기대된다.

Abstract

The tanker market, unlike the container or bulk carrier markets, is dominated by spot transactions, and short-term forecasts play a crucial role in chartering decisions. A hybrid model suggested by Zhang (2003) was applied to forecast tanker freight indices. The model is a combination of the ARIMA and machine learning models i.e. ANN and the SVM. This paper covers not only the BDTI but also the BCTI, which makes it possible to generalize the results. The data length is 240 months, extending from August 1998 to July 2018. The performance test was carried out for the entire period, as well as for two sub-periods (1998-2008 and 2009-2018). The separation of the sub-period was based on the volatility of the index series. Considering spot market decisions, a monthly forecast was attempted. A period of one month is meaningful because spot transactions in the tanker market extend from 30 to 50 days. The results reveal that the hybrid models perform better than single models; they may be used as a guideline to make better decisions in the tanker market, where the share of spot trading is higher than in other sectors.

참고문헌 (37)

  1. 신창훈, "하이브리드 ARIMA-신경망 모델을 통한 컨테이너물동량 예측에 관한 연구", 한국항해항만학회지, 32(2), pp.81~88 (2008), 한국항해항만학회
  2. 신창훈, "항만물동량 예측력 제고를 위한 ARIMA 및 인공신경망모형들의 비교 연구", 한국항해항만학회지, 35(1), pp.83~91 (2011), 한국항해항만학회
  3. 심재희, "계량기법을 이용한 발틱건화물선운임의 예측― 단변량모형, 다변량모형, HP 필터링 기법 ―", 해운물류연구, (58), pp.1~18 (2008), 한국해운물류학회[DOI]
  4. Alizadeh, A. H, "A Regime Switching Approach for Hedging Tanker Shipping Freight Rates", Energy Economics, 49, pp.44~59 (2015)
  5. Atighehchian, A, "An Environment-Driven, Function-Based Approach to Dynamic Single-Machine Scheduling", European Journal of Industrial Engineering, 7(1), pp.100~118 (2013)
  6. Bao, J, "A New BDI Forecasting Model Based on Support Vector Machine", 2016 IEEE Information Technology, Networking, Electronic and Automation Control Conference, pp.65~69 (2016)
  7. Box, G. E. P, "Time Series Analysis: Forecasting and Control" (1976), Holden-Day
  8. Cadenas, E, "Wind Speed Forecasting in Three Different Regions of Mexico, Using a Hybrid ARIMA-ANN Model", Renewable Energy, 35(12), pp.2732~2738 (2010)
  9. Celik, M, "An Integrated Fuzzy QFD Model Proposal on Routing of Shipping Investment Decisions in Crude Oil Tanker Market", Expert Systems with Applications, 36(2), pp.6227~6235 (2009)
  10. Cooke, J, "Voyage Charters" (2014), Routledge
  11. Cybenko, G., "Approximation by Superpositions of a Sigmoidal Function", Mathematics of Control, Signals, and Systems, 2(4), pp.303~314 (1989)
  12. Díaz-Robles, L. A, "A Hybrid ARIMA and Artificial Neural Networks Model to Forecast Particulate Matter in Urban Areas: The Case of Temuco, Chile", Atmospheric Environment, 42(35), pp.8331~8340 (2008)
  13. Eslami, P, "Predicting Tanker Freight Rates Using Parsimonious Variables and a Hybrid Artificial Neural Network with an Adaptive Genetic Algorithm", Maritime Economics & Logistics, 19(3), pp.538~550 (2017)
  14. Fan, S, "Forecasting Baltic Dirty Tanker Index by Applying Wavelet Neural Networks", Journal of Transportation Technologies, 3, pp.68~87 (2013)
  15. Glen, D, "The Handbook of Maritime Economics and Business, edited by Costa Th. Grammenos", pp.355~390 (2010), Lloyd’s List
  16. Han, Q, "Forecasting Dry Bulk Freight Index with Improved SVM", Mathematical Problems in Engineering, 2014, pp.1~12 (2014)
  17. Khashei, M, "A Comparative Study of Series Arima", Mlp Hybrid Models for Stock Price Forecasting, 2018, 1~16
  18. Khashei, M, "An Artificial Neural Network (p, d, q) Model for Time Series Forecasting", Expert Systems with Applications, 37(1), pp.479~489 (2010)
  19. Khashei, M, "A Novel Hybridization of Artificial Neural Networks and ARIMA Models for Time Series Forecasting", Applied Soft Computing Journal, 11(2), pp.2664~2675 (2011)
  20. Kumar, M, "Forecasting Stock Index Returns Using ARIMA-SVM, ARIMA-ANN, and ARIMA-Random Forest Hybrid Models", International Journal of Banking, Accounting and Finance, 5(3), pp.284~308 (2014)
  21. Li, J, "Forecasting Tanker Freight Rate Using Neural Networks", Maritime Policy and Management, 24(1), pp.9~30 (1997)
  22. Lyridis, D, "Forecasting Tanker Market Using Artificial Neural Networks", Maritime Economics &Logistics, 6, pp.93~108 (2004)
  23. Ömer Faruk, D, "A Hybrid Neural Network and ARIMA Model for Water Quality Time Series Prediction", Engineering Applications of Artificial Intelligence, 23(4), pp.586~594 (2010)
  24. OPEC, "World Oil Outlook 2040. Edited by James Griffin and Alvino-Mario Fantini" (2017), The Leading Edge
  25. Pai, P, "A Hybrid ARIMA and Support Vector Machines Model in Stock Price Forecasting", Omega, 33(6), pp.497~505 (2005)
  26. Paliwal, M, "Neural Networks and Statistical Techniques:A Review of Applications", Expert Systems with Applications, 36(1), pp.2~17 (2009)
  27. Santos, A. A. P, "Forecasting Period Charter Rates of VLCC Tankers through Neural Networks: A Comparison of Alternative Approaches", Maritime Economics & Logistics, 16(1), pp.72~91 (2014)
  28. Spreckelsen, C, "Engineering Applications of Neural Networks(EANN 2012) Communications in Computer and Information Science", pp.443~452 (2012), Springer
  29. Stopford, M, "Maritime Economics" (2009), Routledge
  30. Vapnik, V. N, "The Nature of Statistical Learning Theory" (1995), Springer
  31. Vapnik, V. N, "Statistical Learning Theory, New York" (1997), Wiley
  32. Wang, J, "Stock Index Forecasting Based on a Hybrid Model", Omega, 40(6), pp.758~766 (2012)
  33. YANG, Z, "Forecasting Baltic Panamax Index with Support Vector Machine", Journal of Transportation Systems Engineering and Information Technology, 11(3), pp.50~57 (2011)
  34. Zeng, Q, "An Approach for Baltic Dry Index Analysis Based on Empirical Mode Decomposition", Maritime Policy & Management, 41(3), pp.224~240 (2014)
  35. Zeng, Q, "A New Approach for Baltic Dry Index Forecasting Based on Empirical Mode Decomposition and Neural Networks", Maritime Economics & Logistics, 18(2), pp.192~210 (2016)
  36. Zhang, G. P, "Time Series Forecasting Using a Hybrid ARIMA and Neural Network Model", Neurocomputing, 50, pp.159~175 (2003)
  37. Zhang, G, "Forecasting with Artificial Neural Networks:The State of the Art", International Journal of Forecasting, 14(1), pp.35~62 (1998)

관련 논문AI추천

이 간행물의 다른 논문 (8건)