Forecasting World Tuna Catches with ARIMA-Spline and ARIMA-Neural Networks Models

Boonmee  LEE; Suhartono  SUHARTONO; Apiradee  LIM; Sung Keuk  AHN

doi:10.48048/wjst.2021.9726

Authors

Boonmee LEE Department of Mathematics and Computer Science, Faculty of Science and Technology, Prince of Songkla University, Pattani 94000, Thailand https://orcid.org/0000-0001-6502-1169
Suhartono Department of Statistics, Faculty of Mathematics & Natural Sciences, Institut Teknologi Sepuluh Nopember, Surabaya, Indonesia
Apiradee LIM Department of Mathematics and Computer Science, Faculty of Science and Technology, Prince of Songkla University, Pattani 94000, Thailand
Sung Keuk AHN Department of Finance & Management Science, Carson College of Business, Washington State University, WA, USA

DOI:

https://doi.org/10.48048/wjst.2021.9726

Keywords:

ARIMA-spline model, Ensemble forecasts, Neural networks, Nonlinear univariate forecasting, Tuna catches

Abstract

Tuna is a renewable resource that has been managed regionally, but its worldwide capacity for regeneration is still little known. A time-series dataset of tuna catches was used to develop nonlinear univariate models for monitoring the sustainability of tuna catches. Two approaches were compared: 1) fitting an ARIMA-spline model to the volume of annual tuna catches and 2) combining neural networks with an ARIMA model to fit the annual changes in volume. These models offer competitive forecasting performance with small percentage errors. By averaging results of the best model developed in each of these approaches, our ensemble forecast predicts that world tuna catches will reach the optimal level of 5.09 million tons in 2025, remain stable thereafter until 2033, and start decreasing about 0.78 % annually. These models could be used by regional fishery management groups to discover discrepancies between such projections and other science-based estimations of the maximum sustainable output.

HIGHLIGHTS

AnARIMA-spline model is practical for forecasting time series with uncertainties and complex interaction of variables
The plausibility of forecasts is essential as the goodness of fit for statistical model validation
The ensemble forecasts of results from modelling both catches and the changes of catches offer an alternative view for monitoring trend of fishery practices

GRAPHICAL ABSTRACT

Downloads

Download data is not yet available.

Metrics

Metrics Loading ...

References

J Joseph. Managing fishing capacity of the world tuna flee. Fisheries Circular No.982. FAO, Rome, Italy, 2003, p. 23-34.

TK Davies, CC Mees and EJ Milner-Gulland. The past, present and future use of drifting fish aggregating devices (FADs) in the Indian Ocean. Mar. Policy. 2014; 45, 163-70.

Food and Agriculture Organization (FAO). Global production by production source, Available at: http://www.fao.org/fishery/statistics/software/fishstatj/en, accessed March 2017.

A Hamilton, A Lewis, MM McCoy, E Havice and L Campling. Market and industry dynamics in the global tuna supply chain. FFA, Honiara, 2011, p. 18-52.

Food and Agriculture Organization (FAO). Tuna: A global perspective, Available at: http://www.fao.org/docrep/017/ap939e/ap939e.pdf, accessed March 2017.

International Seafood Sustainability Foundation. ISSF tuna stock status update-2016. Washington, D.C., USA, 2016, p. 3-5.

A Fonteneau, E Chassot and N Bodin. Global spatio-temporal patterns in tropical tuna purse seine fisheries on drifting fish aggregating devices (DFADs): Taking a historical perspective to inform current challenges. Aquat. Living Resour. 2013; 26, 37-48.

International Seafood Sustainability Foundation. Management RFMOs, Available at: https://iss-foundation.org/about-tuna/management-rfmos/, accessed March 2017.

J Lenoci. Sustainable tuna fisheries for blue economy. Partnerships in Environmental Management for the Seas of East Asia, Quezon City, Philippines, 2018, p. 4-15.

Western and Central Pacific Fisheries Commission. Summary report: The commission for the conservation and management of highly migratory fish stocks in the Western and Central Pacific Ocean. Thirteenth Regular Session of the Scientific Committee, August 9 - 17. Rarotonga, Cook Islands, 2017, p. 137-9.

E Havice. The structure of tuna access agreements in the Western and Central Pacific Ocean: Lessons for Vessel Day Scheme planning. Mar. Policy 2010; 34, 979-87.

A Harvey. Forecasting, structural time series models and Kalman filter. Cambridge University Press, Cambridge, UK, 1989, p. 14.

JC Gutierrez-Estrada, C Silva, E Yanez, N Rodrıguez and I Pulido-Calvo. Monthly catch forecasting of anchovy Engraulis ringens in the north area of Chile: Non-linear univariate approach. Fish. Res. 2007; 86, 188-200.

L Naranjo, F Plaza, E Yanez, MA Barbieri and F Sanchez. Forecasting of jack mackerel landings (Trachurus murphyi) in central-southern Chile through neural networks. Fish. Oceanogr. 2015; 24, 219-28.

S Chesoh and A Lim. Forecasting fish catches in the Songkhla Lake basin. ScienceAsia 2008; 34, 335-40.

P Komontree, P Tongkumchum and W Karntanut. Trends in marine fish catches at Pattani Fishery Port (1999 - 2003). Songklanakarin J. Sci. Technol. 2006; 28, 887-95.

N McNeil, P Odton and A Ueranantasun. Spline interpolation of demographic data revisited. Songklanakarin J. Sci. Technol. 2011; 33, 117-20.

K Watanabe. In-season forecast of chum salmon return using smoothing spline. Fish. Aquac. J. 2016; 7, 173.

B Lee, D McNeil and A Lim. Spline interpolation for forecasting world tuna catches. In: Proceedings of the International Statistical Institute Regional Statistics Conference: Enhancing Statistics, Prospering Human Life, Bali, Indonesia. 2017, p. 801-7.

N Wongsai, S Wongsai and AR Huete. Annual seasonality extraction using the cubic spline function and decadal trend in temporal daytime MODIS LST data. Remote Sens. 2017; 9, 1254.

B Lee, P Tongkumchum and D McNeil. Forecasting monthly world tuna prices with a plausible approach. Songklanakarin J. Sci. Technol. 2020; 42, 398-405.

Food and Agriculture Organization (FAO). FishstatJ application version 3.02.0: Global capture production 1950 - 2015, Available at: http://www.fao.org/fishery/statistics/software/fishstatj/en, accessed March 2017.

GEP Box and DR Cox. An analysis of transformations. J. Roy. Stat. Soc. Ser. B Methodol. 1964; 26, 211-52.

GEP Box and GM Jenkins. Time series analysis: Forecasting and control. Holden-Day, San Francisco, 1970, p. 47-85.

WWS Wei. Time series analysis: Univariate and multivariate methods. Pearson Education, Boston, 2006, p. 78-103.

JB Ramsey. Tests for specification errors in classical linear least squares regression analysis. J. Roy. Stat. Soc. Ser. B Methodol. 1969; 31, 50-71.

RS Tsay. Nonlinearity tests for times series. Biometrika 1986; 73, 461-6.

TH Lee, H White and CWJ Granger. Testing for neglected nonlinearity in time series models: A comparison of neutral network methods and alternative tests. J. Econom. 1993; 56, 269-90.

R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Available at: http://www.R-project.org, accessed March 2017.

SAS Institute Inc. SAS University Edition, Available at: https://www.sas.com/enus/software/ university-edition.html, accessed June 2017.

WN Venables and BD Ripley. Modern applied statistics with S. Springer, New York, 2002, p. 397-403.

S Wold. Spline functions in data analysis. Technometrics 1974; 16, 1-11.

MA Lukas, FR Hoog and RS Anderssen. Efficient algorithms for robust generalized cross-validation spline smoothing. J. Comput. Appl. Math. 2010; 235, 102-7.

N Molinari, JF Durand and R Sabatier. Bounded optimal knots for regression splines. Comput. Stat. Data Anal. 2004; 45, 159-78.

B Efron and RJ Tibshirani. An Introduction to the Bootstrap. CRC Press LLC., New York, 1993, p. 45-57.

S Lek and JF Guegan. Artificial neural networks as a tool in ecological modeling, an introduction. Ecol. Model. 1999; 120, 65-73.

I Suryanarayana, A Braibanti, RS Rao, VA Ramam, D Sudarsan and GN Rao. Neural networks in fisheries research. Fish. Res. 2008; 92, 115-39.

SS Warren. Neural networks and statistical models. In: Proceedings of the 19th Annual SAS Users Group International Conference, SAS Institute Inc., USA. 1994, p. 1538-50.

GP Zhang. Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing 2003; 50, 159-75.

MS Iyer and RR Rhinehart. A method to determine the required number of neural-network training repetitions. IEEE Trans. Neural Netw. 1999; 10, 427-32.

Food and Agriculture Organization. Global capture production (online query), Available at: http://www.fao.org/fishery/statistics/global-capture-production/query/en, accessed September 2019.

Food and Agriculture Organization. Global capture production (online query), Available at: http://www.fao.org/fishery/statistics/global-capture-production/query/en, accessed May 2021.

Atuna. Industry presses for needed transparency in WCPFC, Available at: http://www.atuna.com/ index.php/en/global-news-registerd/all-atuna-news-registerd, accessed September 2017.

MN Maunder and AE Punt. A review of integrated analysis in fisheries stock assessment. Fish. Res. 2013; 142, 61-74.