FORECASTING WEEKLY SKIPJACK TUNA PRICES IN GENERAL SANTOS CITY USING ARIMAX WITH MONTE CARLO SIMULATION

Document Types

Paper Presentation

Research Theme (for Paper Presentation and Poster Presentation submissions only)

Socio-Economic and Political Landscape (SPL)

School Name

Mindanao State University - General Santos

Track or Strand

Science, Technology, Engineering, and Mathematics (STEM)

Research Advisor (Last Name, First Name, Middle Initial)

Absalon, Juliemer, B.

Start Date

25-6-2026 10:30 AM

End Date

25-6-2026 12:00 PM

Zoom Link/ Room Assignment

Online - https://zoom.us/j/94569671692?pwd=Fj3c3ELOebE6QbqbJOOH9wMuildoEc.1 Meeting ID: 945 6967 1692 | Passcode: research

Abstract/Executive Summary

Price volatility in the weekly skipjack tuna (Katsuwonus pelamis) market in General Santos City challenges fishers, processors, and canneries, necessitating accurate forecasting for informed decision-making and market stability. Using a quantitative descriptive-predictive design, researchers analyzed 181 weekly observations from January 2022 to October 2025, applying ARIMAX modelling with Monte Carlo simulation and the Box-Jenkins method. Initial time series characterization using the Mann-Kendall test, ACF plots, and the ADF test revealed that skipjack tuna prices were stationary, lacked significant seasonality, and showed no persistent trend, while several exogenous variables displayed seasonality and non-stationarity requiring differencing. Incorporating BFAR price data alongside Copernicus Marine Service and Department of Energy variables, the final model utilized Celebes SSHa, Bohol Chl-a, and Celebes Chl-a as exogenous variables, selected through AIC-based lag assessment. The optimal model, ARIMAX(1,0,1), satisfied all diagnostic assumptions and demonstrated solid predictive performance. Training metrics yielded RMSE, MAE, and MAPE of 15.191, 11.286, and 10.35%, respectively, while test values of 20.650, 16.222, and 16.71% indicated only moderate error increase and no severe overfitting. Forecast results suggested a stable two-year price range of ₱107.60–₱109.03 per kilogram. Monte Carlo simulation produced a mean of ₱106.90, a median of ₱105.53, and a wide 95% interval of ₱82.79–₱134.94, reflecting considerable long-term uncertainty. The study concludes that skipjack tuna prices are stationary with no strong persistent trend, and that ARIMAX(1,0,1) effectively captures underlying price dynamics. It recommends exploring nonlinear models, incorporating additional exogenous variables, and leveraging probabilistic forecasts to support procurement planning, inventory management, and policy decision-making.

Keywords

skipjack tuna; ARIMAX modelling; time series analysis; Monte Carlo simulation; forecasting

Statement of Originality

yes

This document is currently not available here.

Share

COinS
 
Jun 25th, 10:30 AM Jun 25th, 12:00 PM

FORECASTING WEEKLY SKIPJACK TUNA PRICES IN GENERAL SANTOS CITY USING ARIMAX WITH MONTE CARLO SIMULATION

Price volatility in the weekly skipjack tuna (Katsuwonus pelamis) market in General Santos City challenges fishers, processors, and canneries, necessitating accurate forecasting for informed decision-making and market stability. Using a quantitative descriptive-predictive design, researchers analyzed 181 weekly observations from January 2022 to October 2025, applying ARIMAX modelling with Monte Carlo simulation and the Box-Jenkins method. Initial time series characterization using the Mann-Kendall test, ACF plots, and the ADF test revealed that skipjack tuna prices were stationary, lacked significant seasonality, and showed no persistent trend, while several exogenous variables displayed seasonality and non-stationarity requiring differencing. Incorporating BFAR price data alongside Copernicus Marine Service and Department of Energy variables, the final model utilized Celebes SSHa, Bohol Chl-a, and Celebes Chl-a as exogenous variables, selected through AIC-based lag assessment. The optimal model, ARIMAX(1,0,1), satisfied all diagnostic assumptions and demonstrated solid predictive performance. Training metrics yielded RMSE, MAE, and MAPE of 15.191, 11.286, and 10.35%, respectively, while test values of 20.650, 16.222, and 16.71% indicated only moderate error increase and no severe overfitting. Forecast results suggested a stable two-year price range of ₱107.60–₱109.03 per kilogram. Monte Carlo simulation produced a mean of ₱106.90, a median of ₱105.53, and a wide 95% interval of ₱82.79–₱134.94, reflecting considerable long-term uncertainty. The study concludes that skipjack tuna prices are stationary with no strong persistent trend, and that ARIMAX(1,0,1) effectively captures underlying price dynamics. It recommends exploring nonlinear models, incorporating additional exogenous variables, and leveraging probabilistic forecasts to support procurement planning, inventory management, and policy decision-making.

https://animorepository.dlsu.edu.ph/conf_shsrescon/2026/BoA_SPL/4