Fourier-Based Modeling of Norway’s Fishing Load Capacity: Nonlinear and Cyclical Dynamics

This study combines the Fourier-augmented Kernel-based regularized least squares (Fourier-KRLS) with the Fourier autoregressive distributed lag (FARDL) to model the fishing load capacity factor (FLCF) in Norway. Traditional linear models often fail to model short-run oscillations and long-run dependencies, as marine resource pressures tend to be nonlinear and cyclical. Our approach corrects this drawback by combining the spectral filtering capability of the Fourier transform with the KRLS model’s resilience to nonlinearities. We discuss the influence of urbanization (URB), foreign direct investment (FDI), and GDP per capita on Norway’s FLCF for 1990–2024. The FARDL model confirms cointegration among FLCF and determinants. URB has a positive long-run effect on FLCF (25.94), while GDP reduces FLCF (−6.92) due to environmental controls and infrastructure. FDI has a weak positive effect (0.11). The speed of adjustment to long-run equilibrium is quick, and the Fourier term values are significant. The Fourier-KRLS output shows that URB has the maximum marginal impact on FLCF. The Fourier terms, statistically significant, show hidden cyclical patterns. The Fourier-KRLS model outperforms the standard KRLS. Our findings prove the applicability of nonlinearity and cycles in modeling environmental sustainability.