Forecasting of tilapia (Oreochromis niloticus) production in Bangladesh using ARIMA model

Abstract

Tilapia farming has expanded rapidly in Bangladesh over the years thanks to a suitable climate for aquaculture and a consistently increasing demand for the fish rich in vitamins and minerals. A clear picture regarding the future trend of tilapia production in Bangladesh is still not available, however. The purpose of this study was to estimate parameters that fit into the Autoregressive Integrated Moving Average (ARIMA) model for forecasting tilapia production in Bangladesh. This was accomplished by calibrating and validating the ARIMA model taking into account the lowest values of the Akaike Information Criterion (AIC) and Bayesian Information Criteria (BIC), graphical arrangements of autocorrelation function (ACF) and partial autocorrelation function (PACF) plots. The best model derived was ARIMA (1, 1, 1), which showed an upward trend of tilapia production since 2006 to date and predicted a similar trend until the year 2040. If this trend continues, the yearly tilapia outturn in the country may reach 690,000 MT, with an upper limit of 1.15 million MT and lower limit of 0.23 million MT, reflecting a substantial increase of around 118% over that produced in 2021. The results of this study will serve as a valuable resource for researchers, decision-makers, academics, and tilapia entrepreneurs, enabling them to develop effective action plans to optimize tilapia production in Bangladesh and strategize for the future integration of tilapia within the country.

Keywords

• Modeling
• ARIMA
• Auto regression
• Moving average
• Autocorrelation function
• Estimating parameters