Modélisation De La Hauteur De Vague Au Sud Et Sud-Est De Madagascar

Henri NIRIKO, Sahoby LALAOHARISOA, Jacques Chrysologue RATSIMAVO, Adolphe RATIARISON

Abstract


The aim of this work is to model climatological mean wave heights in the south and southeast of Madagascar, using the multiple linear regression method within a single model. The explanatory variables considered include wind (meridional and zonal components), oceans currents, sea surface height (SSH), sea surface salinity, sea surface temperature (SST) and atmospheric pressure. A preliminary analysis including the Granger causality test, and the assessment of multicollinearity via the variance inflation factor (VIF) allowed the SSH variable to be eliminated, due to the high value of VIF and the absence of a significant causal link. The final results indicate that the seven remaining explanatory variables effectively predict wave height, with a coefficient of determination (R²) of 0.8916, testifying to the model's strong explanatory capacity.

Keywords


wave height, Granger causality test, multiple linear regression, multicollinearity, VIF.

Full Text:

PDF

References


C. Granger, «Investigating causal relations by econometric models and cross-spectral methods,» Econometrica, p. 424–438, 1969.

M. P. Jacob et K. Eda , «Qu’est-ce que la multicolinéarité ?,» 21 novembre 2023. [En ligne]. Available: https://www.ibm.com/fr-fr/topics/multicollinearity#f15.

J. Smith, L. Johnson et R. Wang, «Oceanic Wind Dynamics and Wave Formation,» Journal of Marine Science, 2020.

A. Dupont, «Climate Regimes and Ocean Conditions,» Climate Research Letters, 2019.

P. Martin et C. Lefèvre, «Statistical Analysis of Wind Effects on Waves,» Oceanography Review, 2021.

R. Kohavi, «A study of cross-validation and bootstrap for accuracy estimation and model selection,» International Joint Conference on Artificial Intelligence, 1995.

S. Arlot et A. Celisse, «A survey of cross-validation procedures,» Statistics Surveys, pp. 40-79, 2010.

T. Hastie, R. Tibshirani et J. Friedman, «The Elements of Statistical Learning: Data Mining, Inference, and Prediction,» Springer, 2009.

C. M. Bishop, «Pattern Recognition and Machine Learning,» Spinger, 2006.

C. David , J. W. Matthijs et J. Giuseppe , «The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation,» PeerJ computer Science, pp. 1-24, 2021.




DOI: http://dx.doi.org/10.52155/ijpsat.v49.1.6948

Refbacks

  • There are currently no refbacks.


Copyright (c) 2025 Niriko Henri Henri

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.