Using Feasible Graphical Lasso Regression Method to Estimate the Parameters of General Linear Regression Model Under High Dimensional Data with Application

The high dimensions problem is one of the important problems that affect the data, and with company of high dimensional data in the general linear regression model, it becomes very difficult to use classical estimation methods such as maximum likelihood method and moments method because these methods lead to biased and inefficient estimates for parameters of the model. The necessity to use new methods to estimate the parameters of the general linear regression model. In this research, two methods were used, the Smooth Integration of Counting Absolute Deviation SICA method, as well as the Feasible Graphical Lasso method FGLasso. Simulation data were used, as well as real data representing the level of pollution in the Tigris River. The results compared using the Mean Square Error MSE and the results showed that the FGLasso estimator is the best