Enhancing monthly lake levels forecasting using heuristic regression techniques with periodicity data component: application of Lake Michigan

This study investigates the accuracy of three different techniques with the periodicity component for estimating monthly lake levels. The three techniques are multivariate adaptive regression splines (MARS), least-square support vector regression (LSSVR), and M5 model tree (M5-tree). Data from Lake Michigan, located in the USA, is used in the analysis. In the first stage of modeling, three techniques were applied to forecast monthly lake level fluctuations up to 8 months ahead of time intervals. In the second stage, the influence of the periodicity component was applied (month number of the year, e.g., 1, 2, 3, horizontal ellipsis 12) as an external subset in modeling monthly lake levels. The root-mean-square error, mean absolute error, and coefficient of determination were used for evaluating the accuracy of the models. In both stages, the comparison results indicate that MARS generally outperforms LSSVR and M5-tree. Further, it has been discovered that including periodicity as an input to the models improves their accuracy in projecting monthly lake levels.

Automated summary

This study investigates the accuracy of three different techniques, MARS, LSSVR, and M5-tree, for estimating monthly lake levels with the periodicity component. The study uses data from Lake Michigan and finds that MARS generally outperforms LSSVR and M5-tree, and including the periodicity component improves the accuracy of the models.