Forecasting weekly salmon prices: Risk management in fish farming

Abstract

Production of farmed salmon has increased substantially during the last decade. Most of the salmon production is sold spot, resulting in large price fluctuations both for the producer and for the exporter. No derivative markets exist; consequently, no one can hedge prices. If prices could be forecasted within reasonable confidence bounds, risk would be reduced. This study used six easily applicable procedures to forecast weekly producer prices for salmon. The procedures tested were Classical Additive Decomposition (CAD), Holt Winters Exponential Smoothing (HW), Auto Regressive Moving Average (ARMA), Vector Auto Regression (VAR) and two different naïve models: post‐sample predictive accuracy was evaluated. Results indicated that the CAD model forecasted the direction of price movements best, whereas the VAR model performed best according to accuracy measures.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 63 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.