Akademik Veri Yönetim Sistemi
AQUACULTURE ECONOMICS AND MANAGEMENT, cilt.30, sa.1, ss.1-17, 2026 (SCI-Expanded, Scopus)
Sea bass and sea bream dominate marine aquaculture production in Türkiye and play a central role in domestic and export markets. Reliable price expectations are therefore essential for producers and traders. This study examines monthly wholesale prices in Türkiye for the period 2016–2025 and evaluates alternative forecasting approaches. Given that both prices are integrated of order one, the Johansen test is applied to assess cointegration. A vector error correction model (VECM) and a robust variant are estimated. Forecast performance is evaluated through rolling-origin backtesting and compared with exponential smoothing, seasonal ARIMA (AutoRegressive Integrated Moving Average), extreme gradient boosting (XGBoost) and a seasonal naïve benchmark using root mean squared error, mean absolute error, mean squared error and mean absolute percentage error. The results show a single cointegrating relationship between sea bass and sea bream prices. Short-run adjustments are asymmetric, with sea bass responding more strongly to disequilibria. In out-of-sample forecasts, VECM-based models produce the lowest error measures. Univariate models perform moderately well, while XGBoost and the seasonal naïve approach yield higher errors. The findings indicate that incorporating long-run equilibrium improves forecast accuracy in integrated seafood markets. Price interdependence between the two species should be considered in applied forecasting and market analysis.