Trends in marine fish production in Tamil Nadu using regression and autoregressive integrated moving average (ARIMA) model
##plugins.themes.bootstrap3.article.main##
Abstract
Tamil Nadu is situated in the south eastern coast of the Indian peninsula with a coastal line of 1076 km (13% of the country’s coast line), 0.19 million sq.km of EEZ (9.4 % of total national EEZ) and a continental shelf of about 41,412 sq. km. This is one of the country’s leading state in marine fish production and ranks third in marine fish production. In Tamil Nadu, Ramanathapuram district is a leading maritime district followed by Nagapattinam and Thoothukudi. The objective of this study was to investigate the trends in marine fish production in Tamil Nadu. Yearly fish production data for the period of 1988-1989 to 2012-2013 were analyzed using time-series method called Autoregressive Integrated Moving Average (ARIMA) model and Regression analysis (curve estimation). In our study, the developed best ARIMA model for Tamil Nadu marine fish production was found to be ARIMA (1, 1, 1) which have the minimum BIC (Bayesian Information Criterion). ARIMA model had got a slightly higher forecasting accuracy rate for forecasting marine fish production of Tamil Nadu than Regression trend analysis. The independent sample test showed there was no significant difference between the two models. The limitations of ARIMA model include its requirement of a long time series data for better forecast. It is basically linear model assuming that data are stationary and have a limited ability to capture non-stationarities and nonlinearities in series data. Both the models indicated that Tamil Nadu marine fish production has plateaued and fishermen should be encouraged to adopt sustainable fishing practices.
##plugins.themes.bootstrap3.article.details##
##plugins.themes.bootstrap3.article.details##
ARIMA, BIC, Marine Production, Sustainable fishing, Trend line regression, Tamil Nadu
Box, G. E. P. and Jenkins, G. M. (1970). Time Series Analysis: Forecasting and Control. San Francisco: Holden Day.
Central Marine Fisheries Research Institute (CMFRI) annual report, 2013- 2014
CMFRI (2010).Marine Fisheries Census, Part II.4, Tamil Nadu
Gutiérrez-Estrada, J. C., Silva, C., Yáñez, E., Rodríguez, N. and Pulido-Calvo, I. (2007). Monthly catch forecasting of anchovy Engraulisringens in the north area of Chile: non-linear univariate approach. Fisheries Research, 86(2-3):188–200
Khashei, M. and Bijari, M. (2010). An artificial neural network (p, d, q) model for time series forecasting. Expert Systems with Applications, 37(1): 479–489
Khashei, M. and Bijari, M. (2011). A new hybrid methodology for nonlinear time series forecasting. Modelling and Simulation in Engineering, 2011, Article ID 379121, 5 pages
Mandal, B. N. (2005). Forecasting sugarcane production in India with ARIMA model, Pp.1-13. Retrieved January 9, 2017 from https://www.researchgate.net/publication/256079794
Srinath, M. and Datta, K. K. (1985). Forecasting marine product exports time-Series analysis. Indian Journal of Fisheries, 32(2): 264-267
Stergiou, K. I., Christou, E. D. and Petrakis, G. (1997). Modelling and forecasting monthly fisheries catches: comparison of regression, univariate and multivariate time series methods. Fisheries Research, 29(1): 55–95
Tamil Nadu State Fisheries Department (TNFD) (2014). ‘Fisheries at a Glance, Marine-2013-14’, Pp. 1-2. Retrieved January 9, 2017 from http://www.fisheries.tn.gov.in/marine-main.html
Tamil Nadu State Fisheries Department (TNFD) (2016). ‘Marine Fisheries Development’, Pp. 1. Retrieved January 9, 2017 from http://www.fisheries.tn.gov.in/marine-main.html
This work is licensed under Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) © Author (s)
