A model for assessing sea environmental quality

In this paper, we aim to study the key factors which affect the environmental quality of the sea. The Fp-ratio, utilized as an indicator of the trophic state of the coastal waters, is classified into three categories. Given the ordinal nature of the response variable, we believe ordinal logistic regression models are the most proper statistic methodology. As the regressors are strongly correlated, the parameters of ordinal logistic regression models cannot be estimated. For solving this problem and, in general, the consequences of multicollinearity, we develop an approach based on principal components. The new approach will be applied for estimating the Fp-ratio.