Higher Dimensional Linear Models: An Application to Shrimp Effort in the Gulf of Mexico (Years 2007-2014) International Journal of Statistics and Applications

The issue of handling categorical along with continuous variables in linear models is always interesting and at the same time challenging. In this paper, a general linear model is extended to include both categorical and continuous predictors. Second order terms including interactions of categorical and continuous variables are discussed. Relations are defined, which map categorical variables onto continuous ones. A special case of such relation is where the levels of categorical variables are nested within the continuous variables hereafter called a “nested” model. The resulting models are then applied to estimate shrimp effort in the Gulf of Mexico for the years 2007 through 2014. A comparison of the results from each of the models is presented in the paper.